Y
Yosi Keller
Researcher at Bar-Ilan University
Publications - 89
Citations - 3494
Yosi Keller is an academic researcher from Bar-Ilan University. The author has contributed to research in topics: Image registration & Motion estimation. The author has an hindex of 26, co-authored 79 publications receiving 2902 citations. Previous affiliations of Yosi Keller include Technion – Israel Institute of Technology & Tel Aviv University.
Papers
More filters
Journal ArticleDOI
A signal processing approach to symmetry detection
Yosi Keller,Yoel Shkolnisky +1 more
TL;DR: It is proved that the AC of symmetric images is a periodic signal whose frequency is related to the order of the symmetry, and this frequency is recovered via spectrum estimation, which is a proven technique in signal processing with a variety of efficient solutions.
Journal ArticleDOI
Spectral Symmetry Analysis
Michael Chertok,Yosi Keller +1 more
TL;DR: The derivation of a symmetry detection and analysis scheme for sets of points IRn and its extension to image analysis by way of local features and improves the scheme's robustness by incorporating geometrical constraints into the spectral analysis.
Journal ArticleDOI
A projection-based extension to phase correlation image alignment
Yosi Keller,Amir Averbuch +1 more
TL;DR: A masking operator is presented that significantly improves the accuracy and robustness of the PC scheme, and is shown to improve the registration of rotated images in the Fourier domain.
Journal ArticleDOI
Kinship verification using multiview hybrid distance learning
Shahar Mahpod,Yosi Keller +1 more
TL;DR: This work proposes a multiview hybrid combined symmetric and asymmetric distance learning network for facial kinship verification, which was successfully applied to the KinFaceW and KinFaceCornell datasets, comparing favorably with contemporary state-of-the-art approaches.
Proceedings ArticleDOI
Fast gradient methods based global motion estimation for video compression
TL;DR: This approach improves existing state-of-the-art GME algorithms by introducing two major modifications: first, only a small subset of the original image pixels is used in the estimation process, which reduces the computational complexity, and second, a warp-free formulation of the basic GM is derived.