Name
Abstract | ||
|---|---|---|
Acquiring and representing the 4D space of rays in the world (the light field) is important for many computer vision and graphics applications. Yet, light field acquisition is costly due to their high dimensionality. Existing approaches either capture the 4D space explicitly, or involve an error-sensitive depth estimation process. This paper argues that the fundamental difference between different acquisition and rendering techniques is a difference between prior assumptions on the light field. We use the previously reported dimensionality gap in the 4D light field spectrum to propose a new light field prior. The new prior is a Gaussian assigning a non-zero variance mostly to a 3D subset of entries. Since there is only a low-dimensional subset of entries with non-zero variance, we can reduce the complexity of the acquisition process and render the 4D light field from 3D measurement sets. Moreover, the Gaussian nature of the prior leads to linear and depth invariant reconstruction algorithms. We use the new prior to render the 4D light field from a 3D focal stack sequence and to interpolate sparse directional samples and aliased spatial measurements. In all cases the algorithm reduces to a simple spatially invariant deconvolution which does not involve depth estimation. |
Year | DOI | Venue |
|---|---|---|
2010 | 10.1109/CVPR.2010.5539854 | Computer Vision and Pattern Recognition |
Keywords | Field | DocType |
deconvolution,image reconstruction,image sequences,rendering (computer graphics),3d focal stack sequence,4d light field spectrum,acquisition techniques,aliased spatial measurements,computer vision,depth invariant reconstruction algorithms,dimensionality gap light field prior,error-sensitive depth estimation process,graphics applications,linear invariant reconstruction algorithms,linear view synthesis,nonzero variance,rendering techniques,sparse directional samples,spatially invariant deconvolution,fourier transforms,estimation,frequency domain analysis,apertures,light field,application software,spectrum,computer graphics,layout,lenses | Iterative reconstruction,Computer vision,Pattern recognition,Computer science,Deconvolution,Curse of dimensionality,Light field,View synthesis,Gaussian,Artificial intelligence,Invariant (mathematics),Rendering (computer graphics) | Conference |
Volume | Issue | ISSN |
2010 | 1 | 1063-6919 |
ISBN | Citations | PageRank |
978-1-4244-6984-0 | 53 | 2.53 |
References | Authors | |
16 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Anat Levin | 1 | 3578 | 212.90 |
| Frédo Durand | 2 | 8625 | 414.94 |