A
Anat Levin
Researcher at Technion – Israel Institute of Technology
Publications - 107
Citations - 14409
Anat Levin is an academic researcher from Technion – Israel Institute of Technology. The author has contributed to research in topics: Speckle pattern & Scattering. The author has an hindex of 42, co-authored 91 publications receiving 12993 citations. Previous affiliations of Anat Levin include Stanford University & Hebrew University of Jerusalem.
Papers
More filters
Proceedings ArticleDOI
Linear view synthesis using a dimensionality gap light field prior
Anat Levin,Frédo Durand +1 more
TL;DR: This paper argues that the fundamental difference between different acquisition and rendering techniques is a difference between prior assumptions on the light field, and proposes a new light field prior which is a Gaussian assigning a non-zero variance mostly to a 3D subset of entries.
Book ChapterDOI
User Assisted Separation of Reflections from a Single Image Using a Sparsity Prior
Anat Levin,Yair Weiss +1 more
TL;DR: This paper focuses on user assisted separation in which the user interactively labels a small number of gradients as belonging to one of the layers, and uses a sparsity prior over derivative filters to achieve good separations from a modest number of labeled gradients.
Journal ArticleDOI
Motion-invariant photography
TL;DR: It is shown that a parabolic integration (corresponding to constant sensor acceleration) leads to motion blur that is invariant to object velocity, and it is proved that the derived parabolic motion preserves image frequency content nearly optimally.
Book ChapterDOI
Patch complexity, finite pixel correlations and optimal denoising
TL;DR: A law of diminishing return is presented, namely that with increasing patch size, rare patches not only require a much larger dataset, but also gain little from it, and this result suggests novel adaptive variable-sized patch schemes for denoising.
Proceedings ArticleDOI
Separating reflections from a single image using local features
TL;DR: This work describes an algorithm that uses an extremely simple form of prior knowledge to perform the decomposition of a single input image into two images that minimize the total amount of edges and comers and shows that this simple prior is surprisingly powerful.