System and method for image-based rendering with object proxies
View Patent ↗A system and method for rendering with an object proxy. In one embodiment, a method includes forming a set of view textures corresponding to a set of viewing directions; selecting a viewing direction for rendering; selecting at least two view textures from the formed set based on the selected viewing direction; and rendering the object proxy at the selected viewing direction. The rendering step includes applying texture from the selected view textures onto the selected object proxy. The view texture set forming step includes: calculating texture coordinates for the object proxy based on the level of obstruction at different portions of the object proxy and texture packing data; and drawing portions of the object based on the level of obstruction data for the object proxy and based on the texture packing data to obtain a view texture at the selected viewing direction.
1. A method for creating a rendering node, the rendering node being useful for drawing an image in a scene at different levels of detail, comprising:
(a) forming a first node component having an object and at least one texture, the first node component configured to draw an image of the object at a first level of detail, wherein the first level of detail has a high resolution;
(b) forming a second node component having an object proxy and a plurality of view textures, the object proxy approximating the geometry of the object, the second node component configured to draw the image of the object at a second level of detail, wherein the second level of detail has an intermediate resolution;
(c) forming a third node component having a billboard with a plurality of views, the third node component configured to draw the image of the object at a third level of detail, wherein the third level of detail has a low resolution; and
(d) combining the first node component, the second node component, and the third node component into a level of detail node, wherein, during rendering, one of the first node component, the second node component, and the third node component of the level of detail node is selected based on an input distance and is used to draw the image of the object.
2. The method of claim 1 , wherein the plurality of view textures are formed by:
(i) selecting a particular viewing direction from a set of viewing directions;
(ii) determining level of obstruction data for each polygon of the object proxy;
(iii) storing the level of obstruction data;
(iv) for each polygon, determining whether the polygon is visible from the selected particular viewing direction and whether texture should be applied from a neighboring viewing direction;
(v) determining texture packing data;
(vi) storing the texture packing data; and
(vii) adjusting texture coordinates of each polygon for the selected particular viewing direction based on said determining step (iv) and said texture packing data.
3. The method of claim 1 , wherein the first level of detail is selected when the input distance indicates that the image of the object is close to a viewer.
4. The method of claim 1 , wherein the third level of detail is selected when the input distance indicates that the image of the object is far from a viewer.
5. The method of claim 1 , further comprising blending at least one of between the first level of detail and the second level of detail and between the second level of detail and the third level of detail.