Abstract: Preservation and reuse of intermediate data generated in a render setup graph for computer animation is disclosed. A processing node in the graph can generate intermediate data and, rather than send it directly to a downstream node in the graph, preserve it for reuse during subsequent processing. As a result, a downstream processing node can reuse the preserved intermediate data, rather than wait while the intermediate data is generated by the processing node in realtime. An intermediate data file management module can manage this process by storing the generated intermediate data in a file for preservation, retrieving the stored intermediate data from the file for reuse, optimizing the file storage location for speed and efficiency, and facilitating sharing of the intermediate data during collaboration between users.

Abstract: A system and method for computing a rendered image of a computer-generated object in a computer-generated scene. A dependency graph is accessed, the dependency graph including a plurality of interconnected nodes including a look-selector node. An asset is accessed at an input to the look-selector node. The asset includes a plurality of looks for the computer-generated object, each look of the plurality of looks corresponding to a different visual appearance of the computer-generated object. At the look-selector node, an active look is selected from the plurality of looks. The active look is passed to a next node of the dependency graph. The rendered image of the computer-generated object is computed having a visual appearance that corresponds to the active look.

Abstract: A lighting correction filter for selectively correcting lighting in computer animation is disclosed. The lighting correction filter can select a computer-generated object having one or more lighting attributes. The selected object can be a portion of an object, an entire object, a portion of a computer-generated scene, or an entire scene. The filter can then set lighting correction values for the lighting attributes of the selected object. The lighting correction values can be color values, exposure values, or both. The filter can apply the lighting correction values to the selected object's lighting attributes to effect a lighting correction in the object prior to rendering.

Abstract: Systems and methods for rendering three-dimensional images by instancing scene description data using a hierarchy are provided. A hierarchy is accessed. The hierarchy comprises a first node and an instance node. The first node is a predecessor to a subtree of one or more nodes and the first node is associated with a first scene description data object. The instance node is a leaf of the hierarchy. The instance node has a parent node and the instance node is associated with a second scene description data object. The parent node has successor nodes other than the instance node. An instancing instruction of the instance node is read. The instancing instruction comprises information identifying the first node. An instance of the subtree of one or more nodes is merged at a location in the hierarchy of the instance node. An image is rendered based on the merged instance of the subtree.

Abstract: A graphical user interface (GUI) for training includes, in some embodiments, a first group of icons arranged about a first axis, where the first group of icons corresponds to computer-generated animation concepts. The GUI also includes a second group of icons arranged about a second axis that intersects the first axis at a particular icon along the first axis. The second group of icons corresponds to videos that illustrate the computer-generated animation concept associated with the particular icon on the first axis. The GUI can also include a third group of icons arranged about a third axis that intersects the first axis at another icon along the first axis. Horizontal correspondence between icons along the second and third axes indicates logical relationships between the corresponding training content.

Abstract: Search-based matching for multiple parameter sets in computer animation is disclosed. The search-based matching method can include receiving a selection of a first set of joint parameters in a first model to match to a second set of joint parameters in a second model, currently in operation, for an appendage of a computer-generated object. The method can also adjust the selected first set of joint parameters to match the second set of joint parameters. The method can further compare the adjusted first set of joint parameters to the second set of joint parameters. Based on the comparison, the method can switch from the second model to the first model and replace the second set of joint parameters with the adjusted first set of joint parameters. The method can then output the replacement first set of joint parameters for rendering the appendage of the object.

Abstract: A system for partitioning a set of assets, where each asset represents a computer-generated object associated with a computer-generated scene. A dependency graph comprising a plurality of interconnected nodes including an organizer node is accessed. The set of assets identified by an input of a predicate test of the organizer node are accessed. It is determined if the at least one predicate test can be evaluated using the set of assets. If the at least one predicate test can be evaluated, one or more partition assets are identified and passed to a next node. If the at least one predicate test cannot be evaluated, a conservative set of assets is identified and passed to the next node, wherein the conservative set of assets is the same set of assets identified by the input of the predicate test.

Abstract: A computer-implemented method for measuring the stereoscopic quality of a computer-generated object in a three-dimensional computer-generated scene. The computer-generated object is visible from at least one camera of a pair of cameras used for creating a stereoscopic view of the computer-generated scene. A set of surface vertices of the computer-generated object is obtained. A stereoscopic transformation on the set of surface vertices is computed to obtain a set of transformed vertices. A translation vector and a scale vector are computed and applied to the set of transformed vertices to obtain a ghosted set of vertices. The ghosted set of vertices is approximately translational and scale invariant with respect to the set of surface vertices. A sum of the differences between the set of surface vertices and the set of ghosted vertices is computed to obtain a first stereo-quality metric.

Abstract: A system for performing graphics processing is disclosed. A dependency graph comprising interconnected nodes is accessed. Each node has output attributes and the dependency graph receives input attributes. A first list is accessed, which includes a dirty status for each dirty output attribute of the dependency graph. A second list is accessed, which associates one of the input attributes with output attributes that are affected by the one input attribute. A third list is accessed, which associates one of the output attributes with output attributes that affect the one output attribute. An evaluation request for a requested output attribute is received. A set of output attributes are selected for evaluation based on being specified in the first list as dirty and being specified in the third list as associated with the requested output attribute. The set of output attributes are evaluated.

Abstract: A computer-implemented method for smoothing a stereo parameter for a computer-animated film sequence. A timeline for the film sequence is obtained, the timeline comprising a plurality of time entries. A stereo parameter distribution is obtained, wherein the stereo parameter distribution comprises one stereo parameter value for at least two time entries of the plurality of time entries, and wherein the stereo parameter value corresponds a stereo setting associated with a pair of stereoscopic cameras configured to produce a stereoscopic image of the computer-animated film sequence. Depending on a statistical measurement of the stereo parameter distribution, either a static scene parameter is calculated, or a set of smoothed parameter values is calculated.

Abstract: Techniques for determining scaled-parallax constraints used for the placement of a pair of stereoscopic cameras within a computer-generated scene. A set of bounded-parallax constraints including a near-parallax value and a far-parallax value is also obtained along with a lower-bound value and upper-bound value for a range of focal lengths. Scaled near-parallax and scaled far-parallax values are calculated, the calculation depending on the whether the focal length is greater than, less than, or within the range of focal lengths.

Abstract: Systems and methods for manipulating a virtual three-dimensional (3D) object in a virtual 3D space are provided. A representation of the 3D object is displayed on a display. A non-hemispherical arcball having a surface is determined. The non-hemispherical arcball is associated with the representation of the 3D object. A pointing device is detected at a first position and at a second position. The first position of the pointing device is translated onto a first location on the surface of the non-hemispherical arcball. The second position of the pointing device is translated onto a second location on the surface of the non-hemispherical arcball. A rotation of the representation of the 3D object is displayed on the display, the rotation based on a path of travel between the first location and the second location along the surface of the non-hemispherical arcball.

Abstract: A computer-implemented method for determining a user-defined stereo effect for a computer-animated film sequence. A stereo-volume value for a timeline of the film sequence is obtained, wherein the stereo-volume value represents a percentage of parallax at the respective time entry. A stereo-shift value for the timeline is also obtained, wherein the stereo-shift value represents a distance across one of: an area associated with a sensor of a pair of stereoscopic cameras adapted to create the film sequence; and a screen adapted to depict a stereoscopic image of the computer-generated scene. A script-adjusted near-parallax value and a script-adjusted far-parallax value are calculated.

Abstract: Systems and methods for performing MOS skin deformations are provided. In one example process, the in vector of a MOS transform may be manually configured by a user. In another example process, a slide/bulge operation may be configured to depend on two or more MOS transforms. Each of the MOS transforms may be assigned a weight that represents the transform's contribution to the overall slide/bulge. In yet another example process, a bulge operation for a MOS vertex may be performed in a direction orthogonal to the attached MOS curve regardless of the direction of the attachment vector. In yet another example process, a ghost transform may be inserted into a MOS closed curve and used to calculate skin deformations associated with first transform of the MOS closed curve.

Abstract: A computer-implemented method for determining bounded-parallax constraints for the placement of a pair of stereoscopic cameras within a computer-generated scene. An initial near-parallax value is determined based on the focal length and a minimum scene depth. An initial far-parallax value is determined based on a focal length. A scaled near-parallax value and scaled far-parallax value are calculated based on the initial near-parallax value, initial far-parallax value, and a range of focal lengths. A creative near-parallax value is calculated based on a stereo-shift value and the product of a stereo-volume and the scaled near-parallax value. A creative far-parallax value is calculated based on the stereo-shift value and the product of the stereo-volume and the scaled far-parallax value. The creative near-parallax value and the creative far-parallax value are stored as the bounded-parallax constraints for the placement of the pair of stereoscopic cameras.

Abstract: Data representing animated hair in a computer generated imagery (CGI) scene may be compressed by treating hair data as arrays of parameters. Hair data parameters may include control vertices, hair color, hair radius, and the like. A principal component analysis (PCA) may be performed on the arrays of hair data. PCA may yield new basis vectors, varying in length, with the largest basis vector corresponding to a new dimension with the largest variance in hair data. The hair data may be quantized based on the varying lengths of new basis vectors. The number of bits allocated for quantizing each new dimension corresponding to each new basis vector may be determined based on the relative lengths of new basis vectors, with more bits allocated to dimensions corresponding to longer basis vectors. The quantized hair data may be bit-packed and then compressed using lossless entropy encoding.

Abstract: To generate a skin-attached element on a skin surface of an animated character, a region of the skin surface within a predetermined distance from a skin-attached element root position is deformed to form a lofted skin according to one of a plurality of constraint surfaces, where each of the plurality of constraint surfaces does not intersect with each other. A sublamina mesh surface constrained to the lofted skin is created. A two-dimensional version of the skin-attached element is projected onto the sublamina mesh surface. The lofted skin is reverted back to a state of the skin surface prior to the deformation of the region of the skin surface.

Abstract: A computer-implemented method for defining a range of bounding parameter values that satisfy perceptual constraints for a stereoscopically filmed computer-generated scene. A user selection of a bounding parameter from a set of scene parameters is selected. Values for scene parameters of the set of scene parameters that were not selected as the bounding parameter are obtained. A first bounding value for the bounding parameter is calculated based on a first perceptual constraint and based on the values of the scene parameters of the set of scene parameters that were not selected. A second bounding value for the bounding parameter is also calculated based on a second perceptual constraint and based on the values of the scene parameter of the set of scene parameters that were not selected. The first and second bounding values define a minimum and a maximum value of a range of values and are stored.

Abstract: A computer-animated scene illuminated by indirect light is shaded. The scene is comprised of sample locations on a surface element of an object in the scene. A point cloud representation of the scene is generated. Optionally, an importance map of the scene, based on the point cloud representation, is generated. The importance map is generated by rasterizing one or more points in the point cloud and designating areas of interest based on the energy value of the one or more points in the point cloud. A ray tracing engine is biased, based on the importance map. The biased ray tracing engine calculates the path of the ray to the sample locations in the scene to an area of interest. The scene is shaded using the output from the biased ray tracing engine.

Abstract: A computer-enabled method for shading locations for use in rendering a computer-generated scene having one or more objects represented by a point cloud. The method involves selecting a shading location, selecting a set of points from the point cloud, rasterizing the points onto a raster shape positioned at the shading location, where the raster shape has varying texel densities that are based on characteristics of the points in the point cloud, such that the texel density varies on different surfaces of the raster shape or on different areas of the same surface or both, and shading the shading location.