ovito.vis

This module contains classes related to data visualization and rendering.

Rendering:

Rendering engines:

Data visualization elements:

Viewport overlays:

class ovito.vis.BondsVis

Base: ovito.vis.DataVis

A visualization element that renders cylindrical bonds between particles. An instance of this class is attached to every Bonds data object and controls the visual appearance of the bonds in rendered images.

See also the corresponding user manual page for this visual element. If you import a simulation file containing bonds, you can subsequently access the BondsVis element through the vis field of the bonds data object, which is part in the data collection managed by the pipeline’s source object:

pipeline = import_file('input/bonds.data.gz', atom_style='bond')
pipeline.add_to_scene()
bonds_vis = pipeline.source.data.particles.bonds.vis
bonds_vis.width = 0.4

In cases where the Bonds data is dynamically generated by a modifier, e.g. the CreateBondsModifier, the BondsVis element is managed by the modifier:

modifier = CreateBondsModifier(cutoff = 2.8)
modifier.vis.flat_shading = True
pipeline.modifiers.append(modifier)
property color

The uniform color of bonds. This parameter is only used if coloring_mode is set to Uniform and if the bonds do not possess a bond property named Color, i.e., explicit per-bond colors were not provided.

Default:

(0.6, 0.6, 0.6)

property coloring_mode

Selects how the color of each bond is determined. Available modes:

  • BondsVis.ColoringMode.Uniform: Use the specified color value to render all bonds.

  • BondsVis.ColoringMode.ByBondType: Use each bond type’s color to render the bonds.

  • BondsVis.ColoringMode.ByParticle: Adopt the colors of the particles connect by the bonds.

Note

If the Bonds object being rendered contains the Color property, then the visual element will directly use these explicit per-bond RGB values for rendering the bonds. The coloring_mode parameter is ignored in this case.

Default:

BondsVis.ColoringMode.ByParticle

property flat_shading

Boolean flag that activates a flat-shaded representation of the bonds instead of the normal cylinder representation.

Default:

False

property width

The display width of bonds (in natural length units).

Default:

0.4

class ovito.vis.ColorLegendOverlay

Base: ovito.vis.ViewportOverlay

This layer renders a color legend over the viewport image, which helps the audience recognize the meaning of depicted object colors. You should add this ViewportOverlay to the Viewport.overlays or Viewport.underlays list of the viewport you use for rendering:

from ovito.vis import ColorLegendOverlay, Viewport
from ovito.qt_compat import QtCore

vp = Viewport(type=Viewport.Type.Top)

legend = ColorLegendOverlay(
    title = 'Potential energy per atom',
    alignment = QtCore.Qt.AlignmentFlag.AlignLeft | QtCore.Qt.AlignmentFlag.AlignTop,
    orientation = QtCore.Qt.Orientation.Vertical,
    offset_y = -0.04,
    font_size = 0.12,
    format_string = '%.2f eV')
vp.overlays.append(legend)

Specifying the color map source

Most importantly, you need to specify which color scheme the legend is supposed to display. There currently are three kinds of color map sources one can use for a legend:

To display the color spectrum of a color coding modifier, including text labels indicating the corresponding numeric value range, set the legend’s modifier field to point to the ColorCodingModifier:

modifier = ColorCodingModifier(property='peatom')
pipeline.modifiers.append(modifier)

legend.modifier = modifier

If you have set up a visual element in your scene that supports pseudo-color mapping, e.g. a VoxelGridVis visualizing some grid property using pseudo-colors, then you can associate it with the color legend by assigning it to the color_mapping_source field:

# Load a VASP volumetric charge density file:
pipeline = import_file('input/CHGCAR.nospin.gz')
pipeline.add_to_scene()

# Configure the VoxelGridVis element responsible for displaying the charge density grid:
grid_vis = pipeline.compute().grids['charge-density'].vis
grid_vis.enabled = True
grid_vis.color_mapping_property = 'Charge density' # Grid property to visualize using pseudo-colors
grid_vis.color_mapping_interval = (0.02186, 0.05465)

# Use the VoxelGridVis as color map source for the legend:
legend.color_mapping_source = grid_vis

To display the discrete colors of a typed Property, specify the source particle or bond property as follows:

legend.property = 'particles/Particle Type'

See the documentation of the property parameter for more information.

property alignment

Selects the corner of the viewport where the color bar is displayed (anchor position). This must be a valid Qt.AlignmentFlag value as shown in the code example above.

Default:

QtCore.Qt.AlignmentFlag.AlignHCenter | QtCore.Qt.AlignmentFlag.AlignBottom

property aspect_ratio

The aspect ratio of the color bar. Larger values make it more narrow.

Default:

8.0

property background_color

The color of the background panel. Only used if background_enabled is set.

Default:

(1.0, 1.0, 1.0)

New in version 3.8.0.

property background_enabled

Enables the painting of a color legend background. A filled panel will be drawn behind the legend to better distinguish the legend from the cluttered 3d scene in the background.

Default:

False

New in version 3.8.0.

property border_color

The line color of the border painted around the color map. This is used only if border_enabled is set.

Default:

(0.0, 0.0, 0.0)

property border_enabled

Enables the painting of a border line around color map.

Default:

False

property color_mapping_source

The DataVis element to be used as color map source by this viewport layer. Set this to the SurfaceMeshVis, VoxelGridVis, TrajectoryVis, or VectorVis element whose color map the legend should display.

Example:

# Add modifier to the pipeline which generates particle trajectory lines:
modifier = GenerateTrajectoryLinesModifier(only_selected=False)
pipeline.modifiers.append(modifier)
modifier.generate()

# Configure the modifier's TrajectoryVis element to apply a color mapping to the lines:
modifier.vis.color_mapping_property = 'Time'
modifier.vis.color_mapping_interval = (0, pipeline.source.num_frames-1)

# Add a color legend and link it to the TrajectoryVis element to display its color map.
vp.overlays.append(ColorLegendOverlay(color_mapping_source=modifier.vis))
Default:

None

New in version 3.7.9.

property font

A string with comma-separated parameter values describing the font to be used for rendering the text labels of the viewport layer. The string must follow the specific form understood by the QFont.fromString() method, for example 'Arial,10,-1,5,75,0,0,0,0,0,Bold'.

Note that the font size parameter (10 in the example specification above) will be ignored by the viewport layer, because the size of text labels is already controlled by the font_size parameter.

property font_size

The relative size of the font used for text labels.

Default:

0.1

property format_string

The format string used with the sprintf() function to generate the text representation of floating-point values. You can change this format string to control the number of displayed decimal places.

Literal text may be incorporated into the format string to include physical units, for example, and HTML text formatting is supported.

Default:

'%g'

property label1

Sets the text string displayed at the upper end of the bar. If empty, the end_value of the ColorCodingModifier is used.

The label supports HTML text formatting.

Default:

''

property label2

Sets the text string displayed at the lower end of the bar. If empty, the start_value of the ColorCodingModifier is used.

The label supports HTML text formatting.

Default:

''

property label_size

The relative size of the font used for the tick labels. Size given relative to font_size.

Default:

0.6

property legend_size

Controls the overall size of the color bar relative to the output image size.

Default:

0.3

property modifier

The ColorCodingModifier for which the color legend should be rendered.

property offset_x

This parameter allows to displace the color bar horizontally from its anchor position. The offset is specified as a fraction of the output image width.

Default:

0.0

property offset_y

This parameter allows to displace the color bar vertically from its anchor position. The offset is specified as a fraction of the output image height.

Default:

0.0

property orientation

Selects the orientation of the color bar. This must be a valid Qt.Orientation value as shown in the code example above.

Default:

QtCore.Qt.Orientation.Horizontal

property outline_color

The text outline color. This is used only if outline_enabled is set.

Default:

(1.0, 1.0, 1.0)

property outline_enabled

Enables the painting of a font outline to make the text easier to read.

Default:

False

property property

Specifies the path to the typed Property for which a discrete color legend should be rendered.

The specified path tells the legend where to find the particle or bond property whose discrete types it should display. Generally, the selected property may be dynamically produced by the current data Pipeline and may not exist yet at the point when you set up the ColorLegendOverlay. That’s why you have to reference it by name instead of specifying the Property object directly.

The path specifies where to find the selected property within the nested containers that make up the DataCollection produced by the current pipeline. It consists of a sequence of DataObject.identifier strings separated by slashes. The last entry in the path is simply the name of the Property to be displayed by the legend.

Examples:

# Display the different structural types identified by PolyhedralTemplateMatchingModifier:
legend.property = 'particles/Structure Type'

# Display the list of bond types in the system:
legend.property = 'particles/bonds/Bond Type'

In case there are multiple data pipelines in the current scene, the legend will automatically pick the first pipeline that yields a DataCollection containing the selected property.

Default:

''

property rotate_title

Enables the vertical orientation of the title, i.e. text orientation parallel to the color legend, if the legend’s orientation is set to QtCore.Qt.Vertical. Otherwise this option is ignored and the title text will be rendered horizontally.

Default:

False

property text_color

The RGB color used for text labels.

Default:

(0.0, 0.0, 0.0)

property ticks_enabled

Enables the painting of tick marks and labels on the color map.

Default:

False

New in version 3.8.0.

property ticks_spacing

Defines the tick spacing along the (continuous) color scale. Requires ticks_enabled to be set. The standard value of 0 activates automatic tick placement based on the input value range.

Default:

0

New in version 3.8.0.

property title

The text displayed next to the color bar. If this string is empty, the title of the source Property object is used.

The title label supports HTML text formatting.

Default:

''

class ovito.vis.CoordinateTripodOverlay

Base: ovito.vis.ViewportOverlay

Displays a coordinate tripod in rendered images. You can attach an instance of this class to a viewport by adding it to the viewport’s overlays collection:

from ovito.vis import CoordinateTripodOverlay, Viewport
from ovito.qt_compat import QtCore

# Create the overlay.
tripod = CoordinateTripodOverlay()
tripod.size = 0.07
tripod.alignment = QtCore.Qt.AlignmentFlag.AlignRight | QtCore.Qt.AlignmentFlag.AlignBottom
tripod.style = CoordinateTripodOverlay.Style.Solid

# Attach overlay to a newly created viewport.
viewport = Viewport(type=Viewport.Type.Perspective, camera_dir=(1,2,-1))
viewport.overlays.append(tripod)
property alignment

Selects the corner of the viewport where the tripod is displayed (anchor position). This must be a valid Qt.AlignmentFlag value value as shown in the example above.

Default:

QtCore.Qt.AlignmentFlag.AlignLeft | QtCore.Qt.AlignmentFlag.AlignBottom

property axis1_color

RGB display color of the first axis.

Default:

(1.0, 0.0, 0.0)

property axis1_dir

Vector specifying direction and length of first axis, expressed in the global Cartesian coordinate system.

Default:

(1.0, 0.0, 0.0)

property axis1_enabled

Enables the display of the first axis.

Default:

True

property axis1_label

Text label for the first axis. Supports formatted text.

Default:

"x"

property axis2_color

RGB display color of the second axis.

Default:

(0.0, 0.8, 0.0)

property axis2_dir

Vector specifying direction and length of second axis, expressed in the global Cartesian coordinate system.

Default:

(0.0, 1.0, 0.0)

property axis2_enabled

Enables the display of the second axis.

Default:

True

property axis2_label

Text label for the second axis. Supports formatted text.

Default:

"y"

property axis3_color

RGB display color of the third axis.

Default:

(0.2, 0.2, 1.0)

property axis3_dir

Vector specifying direction and length of third axis, expressed in the global Cartesian coordinate system.

Default:

(0.0, 0.0, 1.0)

property axis3_enabled

Enables the display of the third axis.

Default:

True

property axis3_label

Text label for the third axis. Supports formatted text.

Default:

"z"

property axis4_color

RGB display color of the fourth axis.

Default:

(1.0, 0.0, 1.0)

property axis4_dir

Vector specifying direction and length of fourth axis, expressed in the global Cartesian coordinate system.

Default:

(0.7071, 0.7071, 0.0)

property axis4_enabled

Enables the display of the fourth axis.

Default:

False

property axis4_label

Label for the fourth axis. Supports formatted text.

Default:

"w"

property font

A string with comma-separated parameter values describing the font to be used for rendering the text labels of the viewport layer. The string must follow the specific form understood by the QFont.fromString() method, for example 'Arial,10,-1,5,75,0,0,0,0,0,Bold'.

Note that the font size parameter (10 in the example specification above) will be ignored by the viewport layer, because the size of text labels is already controlled by the font_size parameter.

property font_size

The font size for rendering the text labels of the tripod. The font size is specified in terms of the tripod size.

Default:

0.4

property line_width

Controls the width of axis arrows. The line width is specified relative to the tripod size.

Default:

0.06

property offset_x

This parameter allows to displace the tripod horizontally. The offset is specified as a fraction of the output image width.

Default:

0.0

property offset_y

This parameter allows to displace the tripod vertically. The offset is specified as a fraction of the output image height.

Default:

0.0

property outline_color

The outline color for text labels. This is used only if outline_enabled is set.

Default:

(1.0, 1.0, 1.0)

property outline_enabled

Enables the painting of a font outline to make the axis labels easier to read.

Default:

False

property size

Scaling factor controlling the overall size of the tripod. The size is specified as a fraction of the output image height.

Default:

0.075

property style

Selects the visual style of the coordinate axis tripod. Supported values are:

  • CoordinateTripodOverlay.Style.Flat (default)

  • CoordinateTripodOverlay.Style.Solid

class ovito.vis.DataVis

Abstract base class for visualization elements that are responsible for the visual appearance of data objects in the visualization. Some DataObjects are associated with a corresponding DataVis element (see DataObject.vis property), making them visual data objects that appear in the viewports and in rendered images.

See the ovito.vis module for the list of visual element types available in OVITO.

property enabled

Boolean flag controlling the visibility of the data. If set to False, the data will not be visible in the viewports or in rendered images.

Default:

True

property title

A custom title string assigned to the visual element, which will show in the pipeline editor of OVITO.

Default:

''

class ovito.vis.DislocationVis

Base: ovito.vis.DataVis

Controls the visual appearance of dislocation lines extracted by a DislocationAnalysisModifier. An instance of this class is attached to every DislocationNetwork data object.

See also the corresponding user manual page for more information on this visual element.

property burgers_vector_color

The color of Burgers vector arrows.

Default:

(0.7, 0.7, 0.7)

property burgers_vector_scaling

The scaling factor applied to displayed Burgers vectors. This can be used to exaggerate the arrow size.

Default:

1.0

property burgers_vector_width

Specifies the width of Burgers vector arrows (in length units).

Default:

0.6

property coloring_mode

Selects the coloring mode for dislocation lines. Supported modes are:

  • DislocationVis.ColoringMode.ByDislocationType (default)

  • DislocationVis.ColoringMode.ByBurgersVector

  • DislocationVis.ColoringMode.ByCharacter

property line_width

Controls the display width (in units of length of the simulation) of dislocation lines.

Default:

1.0

property shading

The shading style used for the lines. Possible values:

  • DislocationVis.Shading.Normal (default)

  • DislocationVis.Shading.Flat

property show_burgers_vectors

Boolean flag that enables the display of Burgers vector arrows.

Default:

False

property show_line_directions

Boolean flag that enables the visualization of line directions.

Default:

False

class ovito.vis.OSPRayRenderer

This is one of the software-based rendering backends of OVITO, which can generate images with higher fidelity than the standard OpenGLRenderer. Typically, you create an instance of this class and pass it to the Viewport.render_image() or Viewport.render_anim() methods.

OSPRay can render scenes with ambient occlusion lighting, semi-transparent objects, and depth-of-field focal blur. For technical details of the supported rendering algorithms and parameters, see the www.ospray.org website. See also the corresponding user manual page for more information on this rendering engine.

property ambient_brightness

Controls the radiance of the ambient light.

Default:

0.8

property ambient_light_enabled

Enables the ambient light, which surrounds the scene and illuminates it from infinity with constant radiance.

Default:

True

property aperture

The aperture radius controls how blurred objects will appear that are out of focus if dof_enabled is set.

Default:

0.5

property denoising_enabled

Enables the application of a denoising filter to the rendered image to reduce Monte Carlo noise inherent to stochastic ray tracing methods like path tracing.

Default:

True

property direct_light_angular_diameter

Specifies the apparent size (angle in radians) of the default directional light source. Setting the angular diameter to a value greater than zero yields soft shadow.

Default:

numpy.radians(10.0)

property direct_light_enabled

Enables the default directional light source that is positioned behind the camera and is pointing roughly along the viewing direction. The brightness of the light source is controlled by the direct_light_intensity parameter.

Default:

True

property direct_light_intensity

The intensity of the default directional light source. The light source must be enabled by setting direct_light_enabled.

Default:

1.0

property dof_enabled

Enables the depth-of-field effect (focal blur). Only objects exactly at the distance from the camera specified by the focal_length will appear sharp when depth-of-field rendering is enabled. Objects closer to or further from the camera will appear blurred. The strength of the effect is controlled by the aperture parameter.

Default:

False

property focal_length

Only objects exactly at this distance from the camera will appear sharp when dof_enabled is set. Objects closer to or further from the camera will appear blurred.

Default:

40.0

property material_shininess

Specular Phong exponent value for the default material. Usually in the range between 2.0 and 10,000.

Default:

10.0

property material_specular_brightness

Controls the specular reflectivity of the default material.

Default:

0.02

property max_ray_recursion

The maximum number of recursion steps during raytracing. Normally, 1 or 2 is enough, but when rendering semi-transparent objects, a larger recursion depth is needed.

Default:

10

property refinement_iterations

The OSPRay renderer supports a feature called adaptive accumulation, which is a progressive rendering method. During each rendering pass, the rendered image is progressively refined. This parameter controls the number of iterations until the refinement stops.

Default:

4

property samples_per_pixel

The number of raytracing samples computed per pixel. Larger values can help to reduce aliasing artifacts.

Default:

2

property sky_albedo

Controls the ground reflectance affecting the sky-sun light source. The light source must be enabled first by setting sky_light_enabled. Valid parameter range is [0.0 - 1.0].

Default:

0.3

property sky_brightness

The intensity of the sky-sun light source. The light source must be enabled first by setting sky_light_enabled.

Default:

2.0

property sky_light_enabled

Enables the sky/sun light source that mimics the light coming from the sky and the sun in an outdoor scene. The brightness of the sky is controlled by the sky_brightness parameter.

Default:

False

property sky_turbidity

Controls atmospheric turbidity due to particles affecting the sky-sun light source. The light source must be enabled first by setting sky_light_enabled. Valid parameter range is [1.0 - 10.0].

Default:

3.0

class ovito.vis.OpenGLRenderer

This is the default rendering backend used by the Viewport.render_image() and Viewport.render_anim() functions. It also serves in the OVITO desktop application for the real-time display of the 3d scene in the interactive viewport windows. The OpenGL renderer uses your computer’s GPU graphics hardware to accelerate the generation of images. See the corresponding user manual page for more information.

Enabling OpenGL support

This rendering backend requires an environment where OpenGL graphics support is available. Standalone Python scripts typically run in a headless environment, e.g. a text-based terminal without graphics support. This prevents the OpenGLRenderer from initializing an OpenGL context and an offscreen framebuffer to render into. A call to Viewport.render_image() will raise an error in this case, and you have to take one of the following steps to enable OpenGL graphics support – or altogether avoid the problem by using one of the software-based rendering backends instead.

The ovito Python module, if imported by a Python script running in an external Python interpreter, sets up a headless environment by default, in which OpenGL rendering is not available. A proper graphics environment can be explicitly requested in two ways (starting with OVITO 3.7.10):

  1. Setting the environment variable OVITO_GUI_MODE=1 prior to importing the ovito package or any of its sub-packages:

    You can set the variable either externally, before or while launching your Python script program:

    OVITO_GUI_MODE=1 python3 <your_script.py>

    or within the Python program itself by including the following lines before the first import ovito statement at the top of your .py file:

    import os
os.environ['OVITO_GUI_MODE'] = '1' # Request a session with OpenGL support

  2. Create a Qt application object with GUI support before importing the ovito module:

    The standard behavior of the ovito module is to create a global QCoreApplication object during module import, which is only suitable for headless operation, e.g., file I/O and pipeline computations, but not sufficient for OpenGL-based rendering. If, however, a global Qt application object already exists at the time the ovito module is loaded, then OVITO will use that pre-existing application object. For example:

    # Make the program run in the context of the graphical desktop environment.
import PySide6.QtWidgets
app = PySide6.QtWidgets.QApplication() 

from ovito.vis import *
vp = Viewport()
vp.render_image(renderer=OpenGLRenderer())

    Note

    Other Python packages imported by your Python script may also be using the Qt cross-platform toolkit (e.g. Matplotlib’s Qt backend). That means such a third-party package may also set up a global Qt application object, which will subsequently be shared with the ovito module. Furthermore, if you are executing your Python script in a graphical IDE such as Spyder, which itself is based on the Qt framework, then a global application instance may already be present at the time the Python script is launched.

Linux/Unix systems (including WSL2)

On Linux/Unix systems, an X11 or Wayland display server is required for OpenGL graphics. When you request a graphics session with one of the methods described above, the Qt framework will attempt to establish a connection to the X11/Wayland server. If this fails, e.g., because the DISPLAY environment variable is not set correctly, Qt reports an error and the program will quit with the following message:

qt.qpa.xcb: could not connect to display 
qt.qpa.plugin: Could not load the Qt platform plugin "xcb" in "" even though it was found.
This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.

Remote servers and HPC clusters, which are typically accessed via SSH terminals, often do not provide a running graphical desktop environment that would allow the use of OpenGL functions by OVITO. In such a headless environment it may still be possible to provide a virtual X server to the Python application, e.g., using the xvfb-run wrapper command:

OVITO_GUI_MODE=1 xvfb-run python3 <your_script.py>
property antialiasing_level

A positive integer controlling the level of supersampling. If 1, no supersampling is performed. For larger values, the image in rendered at a higher resolution and then scaled back to the output size to reduce aliasing artifacts.

Default:

3

class ovito.vis.ParticlesVis

Base: ovito.vis.DataVis

This type of visual element is responsible for rendering particles and is attached to every Particles data object. You can access the element through the vis field of the data object and adjust its parameters to control the visual appearance of particles in rendered images:

from ovito.io import import_file
from ovito.vis import ParticlesVis

pipeline = import_file("input/simulation.dump")
pipeline.add_to_scene()

vis_element = pipeline.compute().particles.vis
vis_element.shape = ParticlesVis.Shape.Square

See also the corresponding user manual page for more information on this visual element.

property radius

The standard display radius of particles. This value is only used if no per-particle or per-type radii have been set. A per-type radius can be set via ParticleType.radius. An individual display radius can be assigned to each particle by setting the Radius particle property, e.g. using the ComputePropertyModifier.

Default:

1.2

property scaling

Global scaling factor that is applied to every particle being rendered.

Default:

1.0

property shape

The kind of shape to use when rendering the particles. Supported modes are:

  • ParticlesVis.Shape.Sphere (default)

  • ParticlesVis.Shape.Box

  • ParticlesVis.Shape.Circle

  • ParticlesVis.Shape.Square

  • ParticlesVis.Shape.Cylinder

  • ParticlesVis.Shape.Spherocylinder

Mode Sphere includes ellipsoid and superquadric particle geometries, which are activated by the presence of the Aspherical Shape and Superquadric Roundness particle properties. Mode Box renders cubic as well as non-cubic boxes depending on the presence of the Aspherical Shape particle property.

Note that this parameter controls the standard shape to be used for all particles. You can override this default setting on a per-particle type basis by setting the ParticleType.shape property to a different value.

class ovito.vis.PythonViewportOverlay

Base: ovito.vis.ViewportOverlay

This type of viewport overlay calls a custom Python script every time an image of the viewport is being rendered. The user-defined script can paint arbitrary graphics on top of the three-dimensional scene.

For more information, see the class ViewportOverlayInterface.

Tip

Instead of using a PythonViewportOverlay it is also possible to directly manipulate the QImage returned by the Viewport.render_image() method before saving the image to disk. A PythonViewportOverlay is only necessary when rendering animations or if you want to use the overlay interactively in the OVITO Pro desktop application.

class Arguments

Data structure passed to legacy render() functions by the system. Modern overlay implementations, which are based on the ViewportOverlayInterface, should NOT use this class. The structure provides access to the painter object, which can be used to issue drawing commands.

Deprecated since version 3.9.1.

property fov

The field of view of the viewport’s camera. For perspective projections, this is the frustum angle in the vertical direction (in radians). For orthogonal projections this is the visible range in the vertical direction (in world units).

property frame

The animation frame number being rendered (0-based).

property is_perspective

Flag indicating whether the viewport uses a perspective projection or parallel projection.

property painter

The QPainter object, which provides painting methods for drawing on top of the image canvas.

property proj_tm

The projection matrix. This 4x4 matrix transforms points from camera space to screen space.

project_point(world_xyz)

Projects a point, given in world-space coordinates, to screen space. This method can be used to determine where a 3d point would appear in the rendered image.

project_size(world_xyz, r)

Projects a size from 3d world space to 2d screen space. This method can be used to determine how large a 3d object, for example a sphere with the given radius r, would appear in the rendered image.

property scene

The current three-dimensional Scene being rendered. Provides access to all visible data pipelines.

property size

A tuple containing the width and height of the viewport image being rendered (in pixels). This may be a sub-region of the output image when rendering a multi-viewport layout.

property view_tm

The affine camera transformation matrix. This 3x4 matrix transforms points/vectors from world space to camera space.

property viewport

The Viewport being rendered.

property delegate

A ViewportOverlayInterface object implementing the logic of the user-defined viewport overlay.

Default:

None

property function

A reference to the Python function to be called every time the viewport is repainted or when an output image is rendered.

The user-defined function must accept exactly one argument as shown in the example above. The system will pass an Arguments object to the function, providing various contextual information on the current frame being rendered.

Default:

None

Deprecated since version 3.9.1.

property pipeline

The Pipeline, which gets automatically evaluated by the system to produce the DataCollection being passed to the ViewportOverlayInterface.render() method during rendering.

When you first insert the PythonViewportOverlay to a viewport’s overlays or underlays lists, the system automatically sets this field to the first data pipeline found in the current visualization Scene.

If your visualization scene contains multiple data pipelines, you should set this field explicitly to select the pipeline whose output data is to used by your overlay.

class ovito.vis.SimulationCellVis

Base: ovito.vis.DataVis

Controls the visual appearance of the simulation cell. An instance of this class is attached to the SimulationCell object and can be accessed through its vis field. See also the corresponding user manual page for this visual element.

The following example script demonstrates how to change the display line width and rendering color of the simulation cell loaded from an input simulation file:

from ovito.io import import_file

pipeline = import_file("input/simulation.dump")
pipeline.add_to_scene()

cell_vis = pipeline.source.data.cell.vis
cell_vis.line_width = 1.3
cell_vis.rendering_color = (0.0, 0.0, 0.8)
property line_width

The width of the simulation cell line (in simulation units of length).

Default:

0.14% of the simulation box diameter

property render_cell

Boolean flag controlling the cell’s visibility in rendered images. If False, the cell will only be visible in the interactive viewports.

Default:

True

property rendering_color

The line color used when rendering the cell.

Default:

(0.0, 0.0, 0.0)

class ovito.vis.SurfaceMeshVis

Base: ovito.vis.DataVis

Controls the visual appearance of a SurfaceMesh object, which is typically generated by modifiers such as ConstructSurfaceModifier or CreateIsosurfaceModifier. See also the corresponding user manual page for more information on this visual element.

property cap_color

The RGB display color of the cap polygons at periodic boundaries.

Default:

(0.8, 0.8, 1.0)

property cap_transparency

The level of transparency of the displayed cap polygons. The valid range is 0.0 – 1.0 (fully opaque to fully transparent).

Default:

0.0

property clip_at_domain_boundaries

Controls whether the mesh gets clipped at non-periodic cell boundaries during visualization. This option plays a role only if the mesh extends beyond the boundaries of the finite domain of the SurfaceMesh; it does not apply to intersections of the surface with periodic boundaries of the simulation domain (see pbc), at which the surface mesh always gets wrapped back into primary cell image for visualization.

If the mesh extends beyond the (non-periodic) domain boundaries, you can use this option to restrict the display of the mesh to those parts that are located inside the domain.

../_images/surface_mesh_vis_clipping_off.png

Clipping off

../_images/surface_mesh_vis_clipping_on.png

Clipping on

Default:

False

New in version 3.8.0.

property color_mapping_gradient

The color gradient for mapping scalar property values taken from the selected color_mapping_property to corresponding RGB color values (color transfer function). See the ColorCodingModifier.gradient parameter for a list of available color gradient types.

Default:

ColorCodingModifier.Rainbow()

property color_mapping_interval

Specifies the range of input values from the selected color_mapping_property getting mapped to corresponding RGB values by the selected color_mapping_gradient. The tuple defines the start and end of the linear interval that is translated to pseudo-colors by the color map. Input property values not within of the interval get mapped to the marginal colors of the selected color map.

Default:

(0.0, 0.0)

property color_mapping_mode

Controls how the color of the surface is computed. Using pseudo-coloring you can visualize a local property of the surface.

Available modes:

  • SurfaceMeshVis.ColorMappingMode.Uniform: Uses surface_color for rendering the entire surface with a constant color (disables local pseudo-coloring).

  • SurfaceMeshVis.ColorMappingMode.Vertex: Colors the surface based on a local property associated with the surface’s vertices.

  • SurfaceMeshVis.ColorMappingMode.Face: Colors the surface based on a local property associated with the surface’s faces.

  • SurfaceMeshVis.ColorMappingMode.Region: Colors the surface based on a local property associated with the surface’s regions.

Default:

SurfaceMeshVis.ColorMappingMode.Uniform

Note

This setting has no effect if the vertices, faces or regions of the SurfaceMesh have explicit colors associated with them, i.e., if the Color property exists in one of these property containers.

property color_mapping_property

The name of the property to be used for coloring the mesh to visualize the local values of this property on the surface. If the Property has several components, then the name must be followed by a component name, e.g. 'Orientation.X'. Whether the property is taken from the vertices, faces, or regions of the SurfaceMesh being rendered is determined by the selected color_mapping_mode.

Numeric values from the source property are mapped to corresponding RGB-based pseudo-colors by first normalizing them according to the specified color_mapping_interval and then applying the selected color_mapping_gradient.

Note that, if the Color property is defined on the surface’s vertices, faces, or regions, then the visual element directly uses these explicit RGB values to render the surface. No color mapping takes place in this case and the color_mapping_property, color_mapping_mode and surface_color parameters are all ignored.

Default:

''

property highlight_edges

Activates the highlighted rendering of the polygonal edges of the mesh.

Default:

False

property reverse_orientation

Flips the orientation of the surface. This affects the generation of cap polygons as well.

Default:

False

property show_cap

Controls the visibility of cap polygons, which are created at the intersection of the surface mesh with the domain boundaries. This option has an effect only if the surface mesh being rendered is closed, which means there are well-defined “interior” and “exterior” regions of space separated by the surface manifold.

Default:

True

See also

SurfaceMeshTopology.is_closed

property smooth_shading

Enables smooth shading of the triangulated surface mesh.

Default:

True

property surface_color

The RGB display color of the surface mesh. Used only if color_mapping_mode is set to uniform coloring.

Default:

(1.0, 1.0, 1.0)

property surface_transparency

The level of transparency of the displayed surface. The valid range is 0.0 – 1.0 (fully opaque to fully transparent).

Default:

0.0

class ovito.vis.TachyonRenderer

This is one of the software-based rendering backends of OVITO. Tachyon is an open-source raytracing engine integrated into OVITO.

An instance of this class can be passed to the Viewport.render_image() or Viewport.render_anim() methods.

Tachyon can render scenes with ambient occlusion lighting, semi-transparent objects, and depth-of-field focal blur. See the corresponding user manual page for more information on this rendering backend.

property ambient_occlusion

Enables ambient occlusion shading. Enabling this lighting technique mimics some of the effects that occur under conditions of omnidirectional diffuse illumination, e.g. outdoors on an overcast day.

Default:

True

property ambient_occlusion_brightness

Controls the brightness of the sky light source used for ambient occlusion.

Default:

0.8

property ambient_occlusion_samples

Ambient occlusion is implemented using a Monte Carlo technique. This parameters controls the number of samples to compute. A higher sample count leads to a more even shading, but requires more computation time.

Default:

12

property antialiasing

Enables supersampling to reduce aliasing effects.

Default:

True

property antialiasing_samples

The number of supersampling rays to generate per pixel to reduce aliasing effects.

Default:

12

property aperture

Controls the aperture of the camera, which is used for depth-of-field rendering.

Default:

0.01

property depth_of_field

This flag enables depth-of-field rendering.

Default:

False

property direct_light

Enables the parallel light source, which is positioned at an angle behind the camera.

Default:

True

property direct_light_intensity

Controls the brightness of the directional light source.

Default:

0.9

property focal_length

Controls the focal length of the camera, which is used for depth-of-field rendering.

Default:

40.0

property shadows

Enables cast shadows for the directional light source.

Default:

True

class ovito.vis.TextLabelOverlay

Base: ovito.vis.ViewportOverlay

Displays a text label in a viewport and in rendered images. You can attach an instance of this class to a viewport by adding it to the viewport’s overlays collection:

from ovito.vis import TextLabelOverlay, Viewport
from ovito.qt_compat import QtCore

# Create the overlay:
overlay = TextLabelOverlay(
    text = 'Some text',
    alignment = QtCore.Qt.AlignmentFlag.AlignHCenter | QtCore.Qt.AlignmentFlag.AlignBottom,
    offset_y = 0.1,
    font_size = 0.03,
    text_color = (0,0,0))

# Attach the overlay to a newly created viewport:
viewport = Viewport(type = Viewport.Type.Top)
viewport.overlays.append(overlay)

Text labels can display dynamically computed values. See the text property for an example.

property alignment

Selects the corner of the viewport where the text is displayed (anchor position). This must be a valid Qt.AlignmentFlag value as shown in the example above.

Default:

QtCore.Qt.AlignmentFlag.AlignLeft | QtCore.Qt.AlignmentFlag.AlignTop

property font

A string with comma-separated parameter values describing the font to be used for rendering the text labels of the viewport layer. The string must follow the specific form understood by the QFont.fromString() method, for example 'Arial,10,-1,5,75,0,0,0,0,0,Bold'.

Note that the font size parameter (10 in the example specification above) will be ignored by the viewport layer, because the size of text labels is already controlled by the font_size parameter.

property font_size

The font size, which is specified as a fraction of the output image height.

Default:

0.02

property format_string

The format string used with the sprintf() function to generate the text representation of global attributes (only floating-point values). You can change this format string to control the number of decimal places shown and switch between exponential and regular notation, for example.

Default:

'%.6g'

property offset_x

This parameter allows to displace the label horizontally from its anchor position. The offset is specified as a fraction of the output image width.

Default:

0.0

property offset_y

This parameter allows to displace the label vertically from its anchor position. The offset is specified as a fraction of the output image height.

Default:

0.0

property outline_color

The text outline color. This is used only if outline_enabled is set.

Default:

(1.0, 1.0, 1.0)

property outline_enabled

Enables the painting of a font outline to make the text easier to read.

Default:

False

property source_pipeline

The Pipeline that is queried to obtain the attribute values referenced in the text string. See the text property for more information.

property text

The text string to be rendered.

The string can contain placeholder references to dynamically computed attributes of the form [attribute], which will be replaced by their actual value before rendering the text label. Attributes are taken from the pipeline output of the Pipeline assigned to the overlay’s source_pipeline property.

The following example demonstrates how to insert a text label that displays the number of currently selected particles:

from ovito.io import import_file
from ovito.vis import TextLabelOverlay, Viewport
from ovito.modifiers import ExpressionSelectionModifier

# Import a simulation dataset and select some atoms based on their potential energy:
pipeline = import_file("input/simulation.dump")
pipeline.add_to_scene()
pipeline.modifiers.append(ExpressionSelectionModifier(expression="peatom > -4.2"))

# Create the overlay. Note that the text string contains a reference
# to an output attribute of the ExpressionSelectionModifier.
overlay = TextLabelOverlay(text="Number of selected atoms: [ExpressionSelection.count]")
# Specify the source of dynamically computed attributes.
overlay.source_pipeline = pipeline

# Attach overlay to a newly created viewport:
viewport = Viewport(type=Viewport.Type.Top)
viewport.overlays.append(overlay)

Tip

You can embed HTML and CSS markup elements in the string to further control the formatting and styling of the text.

Default:

"Text label"

property text_color

The text rendering color.

Default:

(0.0, 0.0, 0.5)

class ovito.vis.TrajectoryVis

Base: ovito.vis.DataVis

Controls the visual appearance of particle trajectory lines. An instance of this class is attached to every TrajectoryLines data object.

You typically create trajectory lines by inserting the GenerateTrajectoryLinesModifier into a data pipeline. The modifier owns a TrajectoryVis element, which is used for visualizing the generated trajectory lines. You can access it through the modifier’s vis field.

property color

The uniform color to be used for rendering the trajectory lines. This parameter is ignored if pseudo-coloring of the lines has been activated by setting a color_mapping_property.

Default:

(0.6, 0.6, 0.6)

property color_mapping_gradient

The color gradient used to map scalar property values from the selected color_mapping_property to corresponding RGB output values (also called color transfer function). See the ColorCodingModifier.gradient parameter for a list of available color gradient types.

Default:

ColorCodingModifier.Rainbow()

property color_mapping_interval

Specifies the range of input values from the selected color_mapping_property getting mapped to corresponding RGB values by the selected color_mapping_gradient. The tuple defines the start and end of the linear interval that is translated to pseudo-colors by the color map. Input property values not within of the interval get mapped to the marginal colors of the selected color map.

Default:

(0.0, 0.0)

property color_mapping_property

The name of the TrajectoryLines property to be used for pseudo-coloring the lines according to the scalar values of this property. If the Property consists of several vector components, then the name must be followed by a specific component name, e.g. 'Velocity.Z'.

Typically, this parameter should be set to the name of the particle property which was sampled during line tracing by the GenerateTrajectoryLinesModifier. See its sample_particle_property parameter for an example.

Numeric values from the selected source property are mapped to corresponding RGB values by first normalizing them according to the specified color_mapping_interval and then applying the selected color_mapping_gradient.

Note

If the TrajectoryLines object being rendered has a property named Color, then this explicit line coloring is used. No color mapping takes place in this case, and the color_mapping_property and color parameters of the visual element are ignored.

Default:

''

property shading

The shading style used for trajectory lines. Possible values:

  • TrajectoryVis.Shading.Normal

  • TrajectoryVis.Shading.Flat (default)

property upto_current_time

If True, trajectory lines are only rendered up to the particle positions at the current animation time. Otherwise, the complete trajectory lines are displayed.

Default:

False

property width

The display width of trajectory lines.

Default:

0.2

property wrapped_lines

If True, the continuous trajectory lines will automatically be wrapped back into the simulation box during rendering. Thus, they will be shown as several discontinuous segments if they cross periodic boundaries of the simulation box.

Default:

False

class ovito.vis.TriangleMeshVis

Base: ovito.vis.DataVis

Controls the visual appearance of a TriangleMesh. See also the corresponding user manual page for more information on this visual element.

property backface_culling

Controls whether triangle faces facing away from the viewer are not rendered.

Default:

False

property color

The uniform RGB color of the triangle mesh, which is used for rendering if the mesh’s faces or vertices have no local colors associated with them. RGB color components must be in the range 0–1.

Default:

(0.85, 0.85, 1.0)

property highlight_edges

Highlights the polygonal edges of the mesh by rendering a wireframe lines along those edges that have been marked as visible.

Default:

False

property transparency

The degree of semi-transparency of the rendered mesh. Valid parameter range is 0.0 – 1.0.

Default:

0.0

class ovito.vis.VectorVis

Base: ovito.vis.DataVis

This kind of visual element renders arrow glyphs for visualizing vectorial data stored in a Property object. See also the corresponding user manual page for more information.

A vector property is any Property array with data type float and three components per element. The VectorVis element supports visualization of vector properties that are stored of the following PropertyContainer types: Particles, Bonds, SurfaceMesh.vertices, SurfaceMesh.faces, and VoxelGrid.

The standard particle properties Force, Displacement, Dipole, and Velocity already have an existing VectorVis element attached to them, which is disabled by default. You must enable it for the arrow glyphs to be displayed, e.g.:

pipeline = import_file('input/simulation.dump')
pipeline.add_to_scene()
vector_vis = pipeline.compute().particles.forces.vis
vector_vis.enabled = True  # This activates the display of arrow glyphs
vector_vis.color = (1,0,0)

In this example, the atomistic forces were loaded as particle property named Force from the imported simulation file, Parameters such as color, width, and flat_shading of the VectorVis element control the visual appearance of the arrow glyphs in rendered images.

Some modifiers in OVITO dynamically add new vector properties to particles. For instance, the CalculateDisplacementsModifier creates the Displacement property and automatically attaches a VectorVis element to it in case you want to visualize the displacements. The visual element is part of the modifier in this case:

modifier = CalculateDisplacementsModifier()
pipeline.modifiers.append(modifier)
modifier.vis.enabled = True  # This activates the display of displacement vectors
modifier.vis.flat_shading = False

If you are writing your own modifier function to compute a vector property, and you want to visualize that property using arrow glyphs, you need to construct a VectorVis element and attach it to the newly created Property object. For example:

def modify(frame, data, vector_vis=VectorVis(alignment=VectorVis.Alignment.Center, color=(1.0, 0.0, 0.4))):

    # Add a new vector property to the particles:
    vector_data = numpy.random.random_sample(size=(data.particles.count, 3))
    property = data.particles_.create_property('My Vector Property', data=vector_data)

    # Attach the visual element to the output property:
    property.vis = vector_vis

Setting up the visual element as an additional parameter of the modify() function provides two advantages: Only a single instance is created, which survives multiple invocations of the modifier function, and OVITO Pro displays the element’s parameter panel in the UI.

property alignment

Controls the positioning of the arrows glyphs with respect to the base points, e.g., the particle positions. Possible values:

  • VectorVis.Alignment.Base (default)

  • VectorVis.Alignment.Center

  • VectorVis.Alignment.Head

property color

The uniform display color of arrow glyphs. This parameter is not used if pseudo-color mapping is enabled through the color_mapping_property option or when the Vector Color property was set for the particles.

Default:

(1.0, 1.0, 0.0)

property color_mapping_gradient

The color gradient used to map scalar property values from the selected color_mapping_property to corresponding RGB output values (color transfer function). See the ColorCodingModifier.gradient parameter for a list of available color gradient types.

Default:

ColorCodingModifier.Rainbow()

property color_mapping_interval

Specifies the range of input values from the selected color_mapping_property getting mapped to corresponding RGB values by the selected color_mapping_gradient. The tuple defines the start and end of the linear interval that is translated to pseudo-colors by the color map. Input property values not within of the interval get mapped to the marginal colors of the selected color map.

Default:

(0.0, 0.0)

property color_mapping_property

The name of a scalar property to be used for coloring the vector glyphs. If the Property has several components, then the name must be followed by a component name, e.g. 'Displacement.Z'.

Numeric values from the selected source property are mapped to corresponding RGB values by first normalizing them according to the specified color_mapping_interval and then applying the selected color_mapping_gradient.

Note that, if a Particles object is being rendered that has a property named Vector Color, then these explicit per-particle colors will be used for rendering the vector glyphs. No color mapping takes place in this case and the color_mapping_property and color parameters of the visual element are ignored.

Default:

''

property flat_shading

Switches between a flat rendering style for the arrows and a three-dimensional representation.

Default:

True

property offset

Additional offset by which all arrows are displaced. This can be used to move the arrows in front of or behind the particles and avoid occlusions.

Default:

(0.0, 0.0, 0.0)

property reverse

Boolean flag which reverves the direction the arrow glyphs.

Default:

False

property scaling

The uniform scaling factor applied to arrow glyphs.

Default:

1.0

property transparency

The level of semi-transparency for rendering the arrows. The valid parameter range is 0.0 – 1.0 (fully opaque to fully transparent).

Default:

0.0

property width

Controls the width of arrows (in simulation units of length).

Default:

0.5

class ovito.vis.Viewport

A viewport is a “window” to the three-dimensional scene, showing the scene from the point of view of a virtual camera.

The virtual camera’s position and orientation are given by the camera_pos and camera_dir properties. Additionally, the type field allows you to switch between perspective and parallel projection modes or reset the camera to one of the standard axis-aligned orientations (top, left, front, etc.). The zoom_all() method repositions the camera automatically such that the entire scene becomes fully visible within the viewport. See also the documentation of the Adjust View dialog of OVITO to learn more about these camera-related settings.

After the viewport’s virtual camera has been set up, you can render an image or movie using the render_image() and render_anim() methods. For example:

from ovito.io import import_file
from ovito.vis import Viewport, TachyonRenderer

pipeline = import_file('input/simulation.dump')
pipeline.add_to_scene()

vp = Viewport(type = Viewport.Type.Ortho, camera_dir = (2, 1, -1))
vp.zoom_all()
vp.render_image(filename='output/simulation.png', 
                size=(320, 240), 
                renderer=TachyonRenderer())

Furthermore, so-called overlays may be added to a viewport. Overlays are function objects that draw additional two-dimensional graphics or text on top of or behind the rendered scene, e.g. coordinate axes or a color legend. See the documentation of the overlays and underlays lists for more information.

property camera_dir

The viewing direction vector of the viewport’s camera.

property camera_pos

The position of the viewport’s camera in the three-dimensional scene.

property camera_up

Direction vector specifying which coordinate axis will point upward in rendered images. Set this parameter to a non-zero vector in order to rotate the camera around the viewing direction and align the vertical direction in rendered images with a different simulation coordinate axis. If set to (0,0,0), then the upward axis is determined by the current user settings set in OVITO’s application settings dialog (z-axis by default).

Default:

(0.0, 0.0, 0.0)

create_jupyter_widget(antialiasing=True, picking=False, vr_scale=0.0, layout=ipywidgets.Layout(width='400px', height='200px')) → ipywidgets.DOMWidget

Creates an interactive widget for embedding in a Jupyter notebook, which displays the 3d scene as seen through this virtual viewport. The method returns an interactive notebook element, which accepts mouse inputs similar to the viewport windows of the OVITO desktop application. It may be necessary to call display in order to show the widget:

vp = Viewport(type=Viewport.Type.Perspective, camera_dir=(0.5, 1.0, -0.4))
vp.zoom_all()
widget = vp.create_jupyter_widget(picking=True)
display(widget)

The Jupyter widget returned by this method is permanently linked to this Viewport instance. Any changes you subsequently make to the non-visual Viewport, for example, setting its camera_pos or camera_dir, will be reflected by the visual viewport widget. Vice versa do all user interactions with the viewport widget update the corresponding fields of the Viewport object.

Important

This method requires the ipywidgets Python package. Please install this package in the Python interpreter used by your Jupyter environment.

Parameters:

  • antialiasing (bool) – Enables multisample anti-aliasing to reduce jagged edges, which appear during WebGL rasterization.

  • picking (bool) – Enables object picking. When hovering the mouse cursor over an object, the widget will display the object’s properties as text.

  • vr_scale (float) – Enables VR support (WebXR browser interface) if set to a positive value. The parameter value specifies the ratio of 1 length unit of the simulation model and 1 meter in VR space. It thus controls the apparent size (scaling) of the model in virtual reality mode. For example, if object dimensions are specified in terms of nanometers in the simulation model, then a vr_scale value of 0.2 would let a 1 nanometer sphere appear 20 centimeters large in virtual reality space.

  • layout (ipywidgets.Layout) – The layout attribute for the new Jupyter widget.

Returns:

ipywidgets.DOMWidget

The layout attribute lets you control the size of the widget, e.g.:

from ipywidgets import Layout
widget = vp.create_jupyter_widget(layout=Layout(width='100%', height='400px'))

Caution

This method is still under active development and not fully functional yet. Expect these (known) limitations:

  • Changes you make to the scene or the viewport camera do not automatically trigger a refresh of the viewport widget. You need to explicitly call widget.refresh() to update the widget display whenever you change the scene.

  • Semi-transparent objects will likely be rendered incorrectly.

  • Viewport layers are not supported yet.

These limitations will be resolved in a future update of the OVITO Python module. Please support the development of this new feature and report any issues you may encounter in our issue tracker or in the OVITO support forum.

New in version 3.7.9.

create_qt_widget(parent=None)

Creates an interactive visual widget displaying the three-dimensional scene as seen through this virtual viewport. The method creates an interactive window accepting mouse inputs from the user similar to the viewport windows of the OVITO desktop application. You can use this method to develop custom user interfaces based on the Qt cross-platform framework that integrate OVITO’s functionality and display the output of a data pipeline.

Parameters:

parent – An optional Qt widget that should serve as parent of the newly created viewport widget.

Returns:

A new QWidget displaying the three-dimensional scene as seen through the virtual viewport.

The Qt widget returned by this method is linked to this Viewport instance. Any changes your Python script subsequently makes to the non-visual Viewport instance, for example setting camera_pos or camera_dir, will automatically be reflected by the visual viewport widget. Vice versa will interactions of the user with the viewport widget automatically lead to changes of the corresponding fields of the Viewport instance.

The following short example program demonstrates the use of the create_qt_widget() method. Please see the Qt for Python documentation for more information on how to create graphical user interfaces using the Qt framework.

See also

Example B4: Standalone trajectory viewer

import sys, os
from PySide6.QtCore import QEventLoop
from PySide6.QtWidgets import QApplication

# Create a global Qt application object - unless we are running inside the 'ovitos' interpreter,
# which automatically initializes a Qt application object.
if not QApplication.instance():
    app = QApplication(sys.argv)

# Note: Import the ovito module AFTER the QApplication object has been created.
# Otherwise, the ovito module automatically creates its own QCoreApplication object,
# which wouldn't let us display a GUI.
from ovito.io import import_file
from ovito.vis import Viewport

# Import model and add it to visualization scene.
pipeline = import_file('input/simulation.dump')
pipeline.add_to_scene()

# Create a virtual Viewport.
vp = Viewport(type=Viewport.Type.Perspective, camera_dir=(2, 1, -1))

# Create a visible GUI widget associated with the viewport.
widget = vp.create_qt_widget()
widget.resize(500, 400)
widget.setWindowTitle('OVITO Viewport Demo')
widget.show()
vp.zoom_all((widget.width(), widget.height()))

# Start the Qt event loop.
if 'app' in locals():
    # Shutdown application when the user closes the viewport widget.
    widget.destroyed.connect(QApplication.instance().quit)
    # When a standard Python interpreter is used to run this script, start the
    # application's main event loop to begin UI event processing.
    sys.exit(app.exec())
else:
    # When running this script with the 'ovitos' interpreter, a Qt event loop is already active.
    # Start a nested event loop then, just for this little demo program.
    eventLoop = QEventLoop()
    widget.destroyed.connect(eventLoop.quit) # Stop event loop when user closes the viewport widget.
    eventLoop.exec()
property fov

The field of view of the viewport’s camera. For perspective projections this is the camera’s angle in the vertical direction (in radians). For orthogonal projections this is the visible range in the vertical direction (in world units).

property overlays

The list of ViewportOverlay objects currently attached to this viewport. Overlays render two-dimensional graphics on top of the three-dimensional scene. See the following overlay types for more information:

To attach a new overlay to the viewport, use the list’s append() method:

from ovito.vis import Viewport, CoordinateTripodOverlay

vp = Viewport(type = Viewport.Type.Ortho)
tripod = CoordinateTripodOverlay(size = 0.07)
vp.overlays.append(tripod)

The viewport also has an underlays list. ViewportOverlay objects inserted into that list will be rendered behind the 3d objects of the scene.

render_anim(filename, size=(640, 480), fps=10, background=(1.0, 1.0, 1.0), renderer=None, range=None, every_nth=1, layout=None, stop_on_error=True)

Renders an animation sequence.

Parameters:

  • filename (str) – The filename under which the rendered animation should be saved. Supported video formats are: .avi, .mp4, .mov and .gif. Alternatively, an image format may be specified (.png, .jpeg). In this case, a series of image files will be produced, one for each frame, which may be combined into an animation using an external video encoding tool of your choice.

  • size – The resolution of the movie in pixels.

  • fps – The number of frames per second of the encoded movie. This determines the playback speed of the animation.

  • background – An RGB triplet in the range [0,1] specifying the background color of the rendered movie.

  • renderer – The rendering engine to use. If none is specified, either OpenGL or Tachyon are used, depending on the availability of OpenGL in the script execution context.

  • range – The interval of frames to render, specified in the form (from,to). Frame numbering starts at 0. If no interval is specified, the entire animation is rendered, i.e. frame 0 through (FileSource.num_frames-1).

  • every_nth – Frame skipping interval in case you don’t want to render every frame of a very long animation.

  • layout – Optional definition of a multi-viewport layout to be rendered into the output image.

  • stop_on_error – Controls whether the rendering process should be aborted by raising an exception in case an error occurs in any of the scene’s data pipelines.

See also the render_image() method for a more detailed discussion of some of these parameters.

render_image(size=(640, 480), frame=0, filename=None, background=(1.0, 1.0, 1.0), alpha=False, renderer=None, crop=False, layout=None, stop_on_error=True)

Renders an image of the viewport’s view.

Parameters:

  • size – A pair of integers specifying the horizontal and vertical dimensions of the output image in pixels.

  • frame (int) – The animation frame to render. Numbering starts at 0. See the FileSource.num_frames property for the number of loaded animation frames.

  • filename (str) – The file path under which the rendered image should be saved (optional). Supported output formats are: .png, .jpeg and .tiff.

  • background – A triplet of RGB values in the range [0,1] specifying the background color of the rendered image.

  • alpha – This option makes the background transparent so that the rendered image may later be superimposed on a different backdrop. When using this option, make sure to save the image in the PNG format in order to preserve the generated transparency information.

  • renderer – The rendering engine to use. If set to None, either OpenGL or Tachyon are used, depending on the availability of OpenGL in the current execution context.

  • crop – This option cuts away border areas of the rendered image filled with the background color; the resulting image may thus turn out smaller than the requested size.

  • layout – Optional definition of a multi-viewport layout to be rendered into the output image. The layout must be provided as a list of Viewport objects and corresponding rectangular areas, which determine where each viewport’s picture appears within the composed output image. Please make use of OVITO Pro’s code generation function to learn how to construct the layout argument.

  • stop_on_error – Controls whether the rendering process should be aborted by raising an exception in case an error occurs in any of the scene’s data pipelines.

Returns:

A QImage object containing the rendered picture.

Populating the scene

Before rendering an image using this method, you should make sure the three-dimensional contains some visible objects. Typically this involves calling the Pipeline.add_to_scene() method on a pipeline to insert its output data into the scene:

pipeline = import_file('simulation.dump')
pipeline.add_to_scene()

Selecting the rendering backend

OVITO supports several different rendering backends for producing pictures of the three-dimensional scene:

Each of these backends exhibits specific parameters that control the image quality and other aspect of the image generation process. Typically, you would create an instance of one of these renderer classes, configure it and pass it to the render_image() method:

vp.render_image(filename='output/simulation.png', 
                size=(320,240),
                background=(0,0,0), 
                renderer=TachyonRenderer(ambient_occlusion=False, shadows=False))

Post-processing images

If the filename parameter is omitted, the method does not save the rendered image to disk. This gives you the opportunity to paint additional graphics on top before saving the QImage later using its save() method:

from ovito.vis import Viewport, TachyonRenderer
from ovito.qt_compat import QtGui

# Render an image of the three-dimensional scene:
vp = Viewport(type=Viewport.Type.Ortho, camera_dir=(2, 1, -1))
vp.zoom_all()
image = vp.render_image(size=(320,240), renderer=TachyonRenderer())

# Paint on top of the rendered image using Qt's drawing functions:
painter = QtGui.QPainter(image)
painter.drawText(10, 20, "Hello world!")
del painter

# Save image to disk:
image.save("output/image.png")

As an alternative to the direct method demonstrated above, you can also make use of a PythonViewportOverlay to paint custom graphics on top of rendered images.

property type

Specifies the projection type of the viewport. The following standard projections are available:

  • Viewport.Type.Perspective

  • Viewport.Type.Ortho

  • Viewport.Type.Top

  • Viewport.Type.Bottom

  • Viewport.Type.Front

  • Viewport.Type.Back

  • Viewport.Type.Left

  • Viewport.Type.Right

The first two types (Perspective and Ortho) allow you to set up custom views with arbitrary camera orientations.

property underlays

The list of ViewportOverlay objects currently attached to this viewport. They render two-dimensional graphics behind the three-dimensional scene. See the overlays list for further information.

zoom_all(size=(640, 480))

Repositions the viewport camera such that all objects in the scene become completely visible. The current orientation (camera_dir) of the viewport’s camera is maintained but the camera_pos and fov parameters are adjusted by this method.

Parameters:

size – Size in pixels of the image that is going to be renderer from this viewport. This information is used to compute the aspect ratio of the viewport rectangle into which the visible objects should be fitted. The tuple should match the size argument being passed to render_image().

Note that this method uses an axis-aligned bounding box computed at frame 0 of the loaded trajectory enclosing all visible objects to adjust the viewport camera. Make sure to call Pipeline.add_to_scene() first to insert some visible object(s) into the scene.

class ovito.vis.ViewportOverlay

Abstract base class for viewport overlays and underlays, which render two-dimensional graphics on top of (or behind) the three-dimensional scene. Examples are CoordinateTripodOverlay, TextLabelOverlay and ColorLegendOverlay. You can also implement your own viewport overlay in Python by using the PythonViewportOverlay class.

property enabled

Controls whether the overlay gets rendered. An overlay can be hidden by setting its enabled property to False.

Default:

True

class ovito.vis.ViewportOverlayInterface

Base: traits.has_traits.HasStrictTraits

Abstract base class for custom viewport overlays written in Python.

When you write a new viewport overlay, you must implement the render() method, which gets called by the system each time a Viewport window is being rendered:

from ovito.vis import ViewportOverlayInterface

class MyOverlay(ViewportOverlayInterface):
    def render(self, canvas: ViewportOverlayInterface.Canvas, **kwargs):
        canvas.draw_text("Hello world", pos=(0.5, 0.5))

If you are working with OVITO Pro, you can add your overlay to one of the interactive viewport windows. If you are writing a standalone Python program to render images or movies using the Viewport.render_image() and Viewport.render_anim() methods, you need to add your overlay to the viewport’s overlays or underlays lists.

When adding your overlay to one of these lists, the instance must be wrapped in a PythonViewportOverlay object as follows:

from ovito.vis import Viewport, PythonViewportOverlay

vp = Viewport()
vp.overlays.append( PythonViewportOverlay( delegate=MyOverlay() ) )

New in version 3.9.2.

class Canvas

The system passes this object to the ViewportOverlayInterface.render() method. It provides various painting functions for drawing 2d graphics on top of (or under) the 3d scene.

draw_image(image, pos=(0.0, 0.0), size=(1.0, 1.0), anchor='south west')

Draws a pixel-based image onto the canvas. The image must be provided as PySide6 QImage. If it includes an alpha channel, the pixel colors will be combined with the canvas’ existing contents using source over blending mode.

The location and size at which to draw the image are both specified in reduced coordinates (relative to the logical_size of the canvas). If you specify None as size, it gets calculated automatically by the method such that each pixel of the image corresponds to one (logical) pixel of the canvas framebuffer. The anchor parameter controls how the image is laid out relative to the anchor position.

Parameters:

  • image – QImage to paint onto the canvas.

  • pos (tuple[float, float]) – Position of the anchor point within the bounds of the canvas in reduced (fractional) coordinates. The origin (0,0) is in the lower left corner of the canvas, (1,1) in the upper right.

  • size (tuple[float, float] | None) – Size of the destination rectangle in reduced (fractional) coordinates.

  • anchor (str) – Position of the anchor point relative to the image. Must be one of "center", "north west", "west", "south west", "south", "south east", "east", "north east", "north".

Tip

The QImage class supports several pixel formats. For best performance, create an QImage using preferred_qimage_format.

Example

from ovito.vis import ViewportOverlayInterface
from ovito.qt_compat import QtGui

class ImageOverlay(ViewportOverlayInterface):
    def render(self, canvas: ViewportOverlayInterface.Canvas, **kwargs):
        image = QtGui.QImage(320, 240, canvas.preferred_qimage_format)
        image.fill(QtGui.QColor.fromRgbF(1.0, 0.0, 0.0, 0.3))
        canvas.draw_image(image, pos=(0.5, 0.5), size=(0.5, 0.5), anchor="center")
draw_text(text, pos, font_size=0.05, anchor='south west', color=(0.0, 0.0, 0.0), alpha=1.0, outline_width=0.0, outline_color=(1.0, 1.0, 1.0), tight_layout=False, rotation=0.0)

Draws a text string onto the canvas.

The location where to draw the text on the canvas is specified in reduced coordinates (relative to the logical_size of the canvas). The anchor parameter controls how the text is laid out relative to the anchor position.

The text string can include HTML markup elements which control the format, e.g., to produce special notations such as superscripts or subscripts. See HTML text formatting for further information.

Parameters:

  • text (str) – The text to draw.

  • pos (tuple[float, float]) – Position of the anchor point on the canvas in reduced (fractional) coordinates. The origin (0,0) is in the lower left corner of the canvas, (1,1) in the upper right.

  • font_size (float) – The size of the text font specified as a fraction of the height of the viewport canvas.

  • anchor (str) – Position of the anchor point relative to the text bounds. This controls the alignment of the text. Must be one of "center", "north west", "west", "south west", "south", "south east", "east", "north east", "north".

  • color (tuple[float, float, float]) – Color of the text. RGB components must be in the range 0-1.

  • alpha (float) – Alpha component of the text color. A value below 1.0 makes the text semi-transparent.

  • outline_width (float) – Width of the outline to draw around the glyphs in units of logical pixels.

  • outline_color (tuple[float, float, float]) – Color of the text outline. RGB components must be in the range 0-1.

  • tight_layout (bool) – Controls whether the true (pixel-precise) bounds of the text are used when laying it out with respect to the anchor position. The default mode is to use a more extended bounding box, which is based on the general ascent and descent (line height) of the font.

  • rotation (float) – Rotation of the text in radians around the anchor point.

See also

text_bounds()

property field_of_view: float

The field of view of the viewport’s camera (Viewport.fov). For perspective projections, this value specifies the frustum angle in the vertical direction (in radians). For orthogonal projections, this is the visible range in the vertical direction (in simulation units of length).

property is_perspective: bool

Indicates whether the 3d view being rendered uses a perspective projection or parallel projection. This depends on the selected Viewport.type.

property logical_size: tuple[int, int]

The width and height in logical pixels of the canvas. This value reflects the output image size set by the user (parameter size of the Viewport.render_image() method).

Note, however, that the logical canvas size may comprise only a sub-region of the full output image when rendering a multi-viewport layout.

mpl_figure(pos=(0.5, 0.5), size=(0.5, 0.5), anchor='center', font_scale=1.0, alpha=0.0, tight_layout=False)

A context manager for creating and rendering a Matplotlib figure onto the canvas.

This method creates a Matplotlib figure and renders it onto the viewport canvas. Inside a with statement, you can add various types of plots to the Matplotlib figure. The figure is then drawn onto the canvas.

Parameters:

  • pos (tuple[float, float]) – Position of the anchor point for placing the figure on the canvas. The origin (0,0) is in the lower left corner of the canvas, (1,1) in the upper right.

  • size (tuple[float, float]) – Size of the figure in relative viewport canvas coordinates, specified as a tuple (width, height).

  • anchor (str) – Position of the anchor point relative to the figure bounds. Must be one of "center", "north west", "west", "south west", "south", "south east", "east", "north east", "north".

  • font_scale (float) – Scaling factor applied to the fonts in the figure. Useful for adjusting the text size of axis labels.

  • alpha (float) – Transparency level of the figure background. A value of 0.0 makes the background fully transparent, while 1.0 makes it opaque.

  • tight_layout (bool) – Controls whether to automatically adjust subplot parameters for a “tight” layout.

Returns:

A generator yielding a Matplotlib figure. The figure is automatically closed and copied to the canvas after the generator exits.

Example

Create a Matplotlib figure, plot some data dynamically calculated by a HistogramModifier as part of a data pipeline, and render it onto the viewport canvas:

from ovito.vis import ViewportOverlayInterface
from ovito.data import DataCollection

class HistogramOverlay(ViewportOverlayInterface):
    def render(self, canvas: ViewportOverlayInterface.Canvas, data: DataCollection, **kwargs):
        with canvas.mpl_figure(pos=(0,1), size=(0.4,0.4), anchor="north west", alpha=1, tight_layout=True) as fig:
            ax = fig.subplots()
            ax.set_title('Potential energy distribution')
            distribution = data.tables['histogram[Potential Energy]'].xy()
            ax.plot(distribution[:,0], distribution[:,1])

Note

To use the mpl_figure() method, you first need to install the matplotlib Python module. See Installing third-party Python modules.

See also

Example O2: Add a data plot to rendered images

property physical_size: tuple[int, int]

The width and height in physical pixels of the canvas onto which the overlay is currently being rendered. This resolution may be higher or lower than the output image size set by the user (logical_size) if rendering currently takes place in an interactive viewport window in the OVITO desktop application.

On the other hand, when rendering an offscreen image, the physical_size typically equals the logical_size – unless some form of supersampling is being performed (i.e., the image is rendered at an increased resolution before it is scaled down to the final output resolution).

property preferred_qimage_format

Optimal QImage.Format to be used with the draw_image() method for best performance. Typically, this will be either Format_ARGB32_Premultiplied or QImage.Format_RGBA8888 depending on the scene renderer currently in use.

project_length(world_xyz, length)

Projects a length or distance from 3d world space to 2d screen space. This method can be used to determine how large a 3d object, for example a sphere with a given radius, would appear in the rendered image.

Additionally to the input length, the method takes a coordinate triplet (x,y,z). This is the location of the base point from where the distance is measured.

Parameters:

  • world_xyz – The (x,y,z) world-space coordinates of the base point.

  • length – The world-space size or distance to be converted to screen space.

Returns:

The computed screen-space size expressed as a fraction of the canvas height.

Return type:

float

project_location(world_xyz)

Projects a point, given in 3d world-space coordinates, to screen space. This method can be used to determine where a 3d point would appear on the canvas in the rendered image.

Note that the projected point may lay outside of the visible viewport region. Furthermore, for viewports with a perspective projection, the input point may lie behind the virtual camera. In this case no corresponding projected point in 2d screen space exists and the method returns None.

Parameters:

world_xyz – The (x,y,z) coordinates of the input point.

Returns:

A (x,y) pair of reduced canvas coordinates; or None if world_xyz is behind the viewer.

Return type:

tuple[float, float] | None

property projection_tm: ndarray

3d projection matrix, which transforms coordinates from camera space to screen space.

qt_painter()

Creates a QPainter object providing advanced drawing methods. The painter lets you paint complex graphics onto the viewport canvas.

The method returns a Pyton context manager, which must be used in a with statement to obtain the actual QPainter:

from ovito.vis import ViewportOverlayInterface
from ovito.qt_compat import QtGui

class PainterOverlay(ViewportOverlayInterface):
    def render(self, canvas: ViewportOverlayInterface.Canvas, **kwargs):

        with canvas.qt_painter() as painter:
            pen = QtGui.QPen(QtGui.QColor(255,0,0))
            brush = QtGui.QBrush(QtGui.QGradient(QtGui.QGradient.OrangeJuice))
            painter.setPen(pen)
            painter.setBrush(brush)
            painter.drawEllipse(painter.window())

The QPainter’s window rect is configured to match the canvas’ logical_size. The QPainter’s viewport rect is configured to match the canvas’ physical_size. See Window-Viewport Conversion.

Internally, the method creates a temporary QImage for the QPainter to draw into. When leaving the with block, that image gets copied to the canvas using the draw_image() method.

text_bounds(text, pos=(0.0, 0.0), font_size=0.05, anchor='south west', outline_width=0.0, tight_layout=False, rotation=0.0)

Calculates the axis-aligned bounding box of a text as if it were drawn by draw_text(). The parameters have the same meaning and default values as in the draw_text() method.

Parameters:

  • text (str) – The text for which to compute the bounding box.

  • pos (tuple[float, float]) – Position of the anchor point on the canvas in reduced (fractional) coordinates. The origin (0,0) is in the lower left corner of the canvas, (1,1) in the upper right.

  • font_size (float) – The size of the text font specified as a fraction of the height of the viewport canvas.

  • anchor (str) – Position of the anchor point relative to the text bounds. This controls the alignment of the text. Must be one of "center", "north west", "west", "south west", "south", "south east", "east", "north east", "north".

  • outline_width (float) – Width of the outline to draw around the glyphs in units of logical pixels.

  • tight_layout (bool) – Controls whether the true (pixel-precise) bounds of the text are used when laying it out with respect to the anchor position. The default mode is to use a more extended bounding box, which is based on the general ascent and descent (line height) of the font.

  • rotation (float) – Rotation of the text in radians around the anchor point.

Returns:

A tuple containing the coordinates of the lower left corner of the bounding box and its size (all in reduced canvas coordinates).

Return type:

tuple[tuple[float, float], tuple[float, float]]

property view_tm: ndarray

Affine transformation matrix encoding the location and orientation of the virtual camera in the three-dimensional scene. This 3 by 4 matrix transforms points/vectors from world space to camera space.

abstract render(canvas, *, data, pipeline, interactive, frame, **kwargs)

Abstract method to implement custom rendering of viewport overlays.

This method must be overridden in subclasses of ViewportOverlayInterface to define the custom rendering behavior for the overlay. When the viewport window is being rendered, this method is called by the system to generate the overlay’s visual content.

Parameters:

  • canvas (Canvas) – An object provided by the system, which offers various painting functions for drawing 2D graphics on top of the 3D scene.

  • data (DataCollection) – The data collection produced by the overlay’s associated pipeline. It contains the dynamically computed data that can be used for rendering the overlay.

  • pipeline (Pipeline | None) – The overlay’s associated pipeline.

  • interactive (bool) – A boolean flag indicating whether the rendering happens in an interactive context (in a GUI environment) or off-screen (e.g., for generating images and animations). Your overlay can use this to perform a long-running computation only in a non-interactive context and render a placeholder instead in the interactive viewports of the OVITO desktop app.

  • frame (int) – The animation frame number currently being rendered.

  • kwargs – Captures any additional arguments that may be passed in by the system.

class ovito.vis.VoxelGridVis

Base: ovito.vis.DataVis

This visual element controls the appearance of a VoxelGrid data object, which is typically generated by the SpatialBinningModifier or imported directly from files containing volumetric data. The visual element is responsible for rendering the outer boundaries of the grid, i.e., showing only the voxel cells at the surface but not in the interior of the grid volume.

See also the corresponding user manual page for further information on this visual element.

property color_mapping_gradient

The color gradient used to map scalar property values from the selected color_mapping_property to corresponding RGB output values (also called color transfer function). See the ColorCodingModifier.gradient parameter for a list of available color gradient types.

Default:

ColorCodingModifier.Rainbow()

property color_mapping_interval

Specifies the range of input values from the selected color_mapping_property getting mapped to corresponding RGB values by the selected color_mapping_gradient. The tuple defines the start and end of the linear interval that is translated to pseudo-colors by the color map. Input property values not within of the interval get mapped to the marginal colors of the selected color map.

Default:

(0.0, 0.0)

property color_mapping_property

The name of the VoxelGrid scalar property to be used for coloring the grid cells. If the Property has several components, then the name must be followed by a component name, e.g. 'Velocity.Z'.

Numeric values from the selected source property are mapped to corresponding RGB cell colors by first normalizing them according to the specified color_mapping_interval and then applying the selected color_mapping_gradient.

Note that, if the VoxelGrid being rendered contains the Color property, then the visual element directly uses these RGB values to render the grid cells. No color mapping takes place in this case and the color_mapping_property parameter is ignored.

Default:

''

property highlight_grid_lines

Controls the rendering of grid lines separating the voxel cells.

Default:

True

property interpolate_colors

Controls whether the (pseudo)colors of the voxel cells visible on the boundary of the grid are smoothly interpolated between neighboring cells.

Default:

False

property transparency

The level of transparency of the displayed grid surface. The valid parameter range is 0.0 – 1.0 (fully opaque to fully transparent).

Default:

0.0