S
Shai Avidan
Researcher at Tel Aviv University
Publications - 153
Citations - 17052
Shai Avidan is an academic researcher from Tel Aviv University. The author has contributed to research in topics: Pixel & Computer science. The author has an hindex of 50, co-authored 138 publications receiving 15378 citations. Previous affiliations of Shai Avidan include Mitsubishi Electric Research Laboratories & Mitsubishi.
Papers
More filters
Journal ArticleDOI
Underwater Single Image Color Restoration Using Haze-Lines and a New Quantitative Dataset
TL;DR: This work places multiple color charts in the scenes and calculated its 3D structure using stereo imaging to obtain ground truth, and contributes a dataset of 57 images taken in different locations that enables a rigorous quantitative evaluation of restoration algorithms on natural images for the first time.
Proceedings ArticleDOI
Object retrieval and localization with spatially-constrained similarity measure and k-NN re-ranking
TL;DR: A new spatially-constrained similarity measure (SCSM) is proposed to handle object rotation, scaling, view point change and appearance deformation, and a novel and robust re-ranking method with the k-nearest neighbors of the query for automatically refining the initial search results.
Proceedings ArticleDOI
Coherency Sensitive Hashing
Simon Korman,Shai Avidan +1 more
TL;DR: Coherency Sensitive Hashing is verified on a new, large scale, data set of 133 image pairs and is at least three to four times faster than PatchMatch and more accurate, especially in textured regions, where reconstruction artifacts are most noticeable to the human eye.
Proceedings ArticleDOI
Synthetic Aperture Tracking: Tracking through Occlusions
TL;DR: This system is the first capable of tracking in the presence of such significant occlusion, and does not require explicit modeling or reconstruction of the scene and enable tracking in complex, dynamic scenes with moving cameras.
Book ChapterDOI
Multiple hypothesis video segmentation from superpixel flows
TL;DR: This work determines the solution of this segmentation problem as the MAP labeling of a higher-order random field, and develops a framework that allows MHVS to achieve spatial and temporal long-range label consistency while operating in an on-line manner.