Multi-perspective stereoscopy from light fields
Methods and systems for generating stereoscopic content with granular control over binocular disparity based on multi-perspective imaging from representations of light fields are provided. The stereoscopic content is computed as piecewise continuous cuts through a representation of a light field, minimizing an energy reflecting prescribed parameters such as depth budget, maximum binocular disparity gradient, desired stereoscopic baseline. The methods and systems may be used for efficient and flexible stereoscopic post-processing, such as reducing excessive binocular disparity while preserving perceived depth or retargeting of already captured scenes to various view settings. Moreover, such methods and systems are highly useful for content creation in the context of multi-view autostereoscopic displays and provide a novel conceptual approach to stereoscopic image processing and post-production.
1. A method of deferred rendering of computer-generated content, the method comprising:
generating a respective depth map for each of a set of computer-generated input images of a scene;
generating a disparity volume from the generated depth maps, the disparity volume corresponding to a three-dimensional (3D) representation of a light field representing light rays from the scene;
determining a non-planar surface cut through the disparity volume, wherein the non-planar surface cut is determined using a two-dimensional map;
determining a plurality of two-dimensional (2D) subsets of the disparity volume corresponding to a plurality of 2D subsets of the 3D representation of the light field, wherein at least one 2D subset of the disparity volume corresponds to the non-planar surface cut;
determining a plurality of the computer-generated input images whose respective depth maps are included in the determined 2D subsets of the disparity volume; and
rendering, by a computer system, the determined plurality of the computer-generated input images.
2. The method of claim 1 , further comprising generating the plurality of 2D subsets of the 3D representation of the light field.
3. The method of claim 1 , further comprising:
determining a set of the light rays from the scene included in the plurality of 2D subsets of the disparity volume; and
rendering, by a computer system, the determined set of light rays.
4. The method of claim 1 , further comprising generating a stereoscopic image pair of the scene using the rendered plurality of the computer-generated input images.
5. The method of claim 1 , wherein the set of computer-generated input images of the scene comprises perspective images of the scene at linearly translated camera positions.
6. The method of claim 1 , further comprising rendering the depth maps.
7. The method of claim 6 , wherein the disparity volume is generated using the rendered depth maps.
8. A system comprising:
one or more processors configured to:
generate a respective depth map for each of a set of computer-generated input images of a scene;
generate a disparity volume from the generated depth maps, the disparity volume corresponding to a three-dimensional (3D) representation of a light field representing light rays from the scene;
determine a non-planar surface cut through the disparity volume, wherein the non-planar surface cut is determined using a two-dimensional map;
determine a plurality of two-dimensional (2D) subsets of the disparity volume corresponding to a plurality of 2D subsets of the 3D representation of the light field, wherein at least one 2D subset of the disparity volume corresponds to the non-planar surface cut;
determine a plurality of the computer-generated input images whose respective depth maps are included in the determined 2D subsets of the disparity volume; and
render, by a computer system, the determined plurality of the computer-generated input images.
9. The system of claim 8 , wherein the processors are further configured to generate the plurality of 2D subsets of the 3D representation of the light field.
10. The system of claim 8 , wherein the processors are further configured to:
determine a set of the light rays from the scene included in the plurality of 2D subsets of the disparity volume; and
render, by a computer system, the determined set of light rays.
11. The system of claim 8 , wherein the processors are further configured to generate a stereoscopic image pair of the scene using the rendered plurality of the computer-generated input images.
12. The system of claim 8 , wherein the set of computer-generated input images of the scene comprises perspective images of the scene at linearly translated camera positions.
13. The system of claim 8 , wherein the processors are further configured to render the depth maps.
14. The system of claim 13 , wherein the disparity volume is generated using the rendered depth maps.
15. A non-transitory computer-readable medium containing program code that, when executed by a computer, causes the computer to perform a method of deferred rendering of computer-generated content comprising:
generating a respective depth map for each of a set of computer-generated input images of a scene;
generating a disparity volume from the generated depth maps, the disparity volume corresponding to a three-dimensional (3D) representation of a light field representing light rays from the scene;
determining a non-planar surface cut through the disparity volume, wherein the non-planar surface cut is determined using a two-dimensional map;
determining a plurality of two-dimensional (2D) subsets of the disparity volume corresponding to a plurality of 2D subsets of the 3D representation of the light field, wherein at least one 2D subset of the disparity volume corresponds to the non-planar surface cut;
determining a plurality of the computer-generated input images whose respective depth maps are included in the determined 2D subsets of the disparity volume; and
rendering, by a computer system, the determined plurality of the computer-generated input images.
16. The non-transitory computer-readable medium of claim 15 , the method further comprising:
determining a set of the light rays from the scene included in the plurality of 2D subsets of the disparity volume; and
rendering, by the computer, the determined set of light rays.
17. The non-transitory computer-readable medium of claim 15 , the method further comprising generating a stereoscopic image pair of the scene using the rendered plurality of the computer-generated input images.
18. The non-transitory computer-readable medium of claim 15 , wherein the set of computer-generated input images of the scene comprises perspective images of the scene at linearly translated camera positions.
19. The non-transitory computer-readable medium of claim 15 , the method further comprising rendering the depth maps.
20. The non-transitory computer-readable medium of claim 19 , wherein the disparity volume is generated using the rendered depth maps.