Sprite compositor and method for performing lighting and shading operations using a compositor to combine factored image layers

Claim 3

The method of claim 1 wherein two or more of the layers are rendered at different spatial resolution.

Claim 4

The method of claim 1 further including computing a 2D image warp from changes in position of characteristic points of a 3D object in the 3D scene between a previous frame and a subsequent frame and warping a rendered image layer from the previous frame to approximate change in the image layer from the previous frame to the subsequent frame.

Claim 5

The method of claim 1 wherein the method further includes computing a sprite transform for each image layer by projecting 3D object geometry into screen coordinates, finding a bounding region for the projection of the 3D object in the screen coordinates, and deriving the sprite transform from the bounding region; storing the sprite transforms for each of the corresponding image layers, where each sprite transform defines a mapping of an image layer from sprite coordinates to the screen coordinates, and 3D geometry corresponding to the image layer is rendered into a 2D space defined by the sprite coordinates to create the image layer.

Claim 6

The method of claim 4 including updating the sprite transforms of two or more of the image layers at different rates than the corresponding image layers are updated.

Claim 7

The method of claim 4 wherein the sprite transform of at least one of the image layers is used to transform an associated image layer rendered to a first spatial resolution to a second, different spatial resolution than the first spatial resolution.

Claim 8

The method of claim 1 wherein the image operator includes an over image operator.

Claim 9

The method of claim 1 wherein the image operator includes an add image operator for adding first and second image layers together.

Claim 10

The method of claim 1 wherein the image operator includes a multiply image operator for multiplying a first image layer by an array of coefficients.

Claim 11

The method of claim 1 wherein the step of combining image layers includes using two or more image operators.

Claim 12

The method of claim 10 wherein the two or more image operators include an over image operator and an add image operator.

Claim 13

The method of claim 11 wherein the two more image operators further include the multiply image operator.

Claim 14

The method of claim 1 wherein the shading model is factored into first and second lighting terms, and one of the image layers corresponding to the first lighting term represents a rendering of a 3D object illuminated by a first light source and another one of the image layers corresponding to the second lighting term represents a rendering of the 3D object illuminated by a second light source; and further including: rendering the 3D object illuminated by the first light source to a first image layer; rendering the 3D object illuminated by the second light source to a second image layer; and compositing the first and second image layers into the output image.

Claim 15

The method of claim 1 wherein the shading model is factored into first and second texture layers, and one of the image layers corresponding to the first texture layer represents a first texture mapped to 3D object geometry in the scene and another image layer corresponding to the second texture layer represents a second texture mapped to the 3D object geometry; and further including: rendering the 3D object geometry, including texture mapping a first texture image to the 3D object geometry, to compute a first image layer; rendering the 3D object geometry, including texture mapping a second texture image to the 3D object geometry, to compute a second image layer; and compositing the first and second image layers into the output image.

Claim 16

The method of claim 1 wherein one of the image layers represents a shadow layer in the 3D scene, where the scene has a viewpoint and a light source, and rendering the shadow layer includes: rendering the 3D scene from the perspective of the light source to generate a depth map; and rendering the 3D scene from the perspective of the view point to determine shadow attenuation coefficients using the depth map; wherein the shadow layer is re-rendered at a different rate than a rendered image layer of 3D object geometry to which the shadow layer is applied.

Claim 17

The method of claim 1 wherein one of the image layers represents a reflection layer, and further including rendering a 3D object in the scene into the reflection layer and re-rendering the 3D object to the reflection layer at a different rate than other layers.

Claim 20

The method of claim 17 further including warping a rendered image layer of a 3D object from a previous frame to approximate change in lighting or geometric position of the object represented in the image layer from the previous frame to a subsequent frame.

Claim 21

The method of claim 18 wherein the warping step comprises computing a color warp by sampling a lighting model at characteristic points of a 3D object in the previous frame and the subsequent frame to compute lighting changes, and finding a color warp that approximates the lighting changes at the characteristic points; and performing the color warp on the rendered image layer from the previous frame to approximate the change in color values from the previous to the subsequent frame.

Claim 22

The method of claim 18 further including computing a 2D image warp from changes in position of characteristic points of a 3D object in the 3D scene between the previous frame to the subsequent frame; wherein the warping step comprises performing the 2D image warp to transform the rendered image layer from a first screen position for the previous frame to a second screen position in the subsequent frame.

Claim 23

The method of claim 17 wherein the method further includes storing a sprite transform with each of the image layers, where the sprite transform defines a mapping of an image layer from sprite coordinates to screen coordinates, and 3D geometry corresponding to the image layer is rendered into a 2D space defined by the sprite coordinates to create the image layer.

Claim 24

The method of claim 21 including updating the sprite transforms of two or more of the image layers at different rates than the corresponding image layers are updated.

Claim 25

A computer-readable medium having computer executable instructions for performing the steps of claim 17.

Claim 26

A computer-readable medium having computer executable instructions for performing the steps of claim 1.

Claim 27

A sprite compositor for performing shading operations in a graphics rendering system comprising: a set of buffers for storing pixel values from factored image layers, where the image layers correspond to terms in a shading model, where the shading model represents a description of shading to be applied to 3D scene geometry, the shading model is factored over a 3D object or set of 3D objects in a 3D scene, and the 3D object or set of 3D objects are independently and separately rendered to each of the image layers such that a different term in the shading model applies for each independent rendering to the layers; an input multiplexer for selecting input pixels from among the set of buffers; compositing logic including at least one image operator for combining input pixels from separate image layers; a demultiplexer coupled to the compositing logic for receiving combined input pixels and selecting one of the set of buffers to store the combined input pixels; and an output multiplexer for selecting one of the set of buffers to transfer display pixels from the sprite compositor.

Claim 28

A sprite compositor for performing shading operations in a graphics rendering system comprising: a set of buffers for storing pixel values from factored image layers, where the image layers correspond to terms in a shading model, where the shading model represents a description of shading to be applied to 3D scene geometry, the shading model is factored over a 3D object or set of 3D objects in a 3D scene, and the 3D object or set of 3D objects are independently and separately rendered to each of the image layers such that a different term in the shading model applies for each independent rendering to the layers; an input multiplexer for selecting input pixels from among the set of buffers; compositing logic including at least one image operator for combining input pixels from separate image layers; a demultiplexer coupled to the compositing logic for receiving combined input pixels and selecting one of the set of buffers to store the combined input pixels; and an output multiplexer for selecting one of the set of buffers to transfer display pixels from the sprite compositor.

Claim 29

The sprite compositor of claim 25 wherein the compositing logic includes two or more image operators for combining input pixels from separate image layers into output pixels of an output image.

Claim 30

The sprite compositor of claim 26 wherein the compositing logic includes a multiply image operator for multiplying input pixels by modulation factors from a separate image layer.

Claim 33

The method of claim 28 wherein at least one of the layers is rendered at a different spatial resolution than the output image and then scaled to the spatial resolution of the output image for display.

Claim 36

The method of claim 30 further including: factoring a shading model of a 3D object in the scene geometry into separate layers, where one or more of the separate layers of the factored shading model have a blending factor; in response to the user editing the blend factor of an edited layer, rendering the edited layer of the 3D object associated with the blend factor and combining the edited layer with the pre-rendered layers such that the 3D rendering resources are allocated to the edited layer.