• Publications
  • Influence
Non-homogeneous Content-driven Video-retargeting
The proposed algorithm is fully automatic and based on local saliency, motion detection and object detectors, and compared to the state of the art in image retargeting.
The Action Similarity Labeling Challenge
This paper presents a novel video database, the “Action Similarity LAbeliNg” (ASLAN) database, along with benchmark protocols, and makes the ASLAN database, benchmarks, and descriptor encodings publicly available to the research community.
Optimizing Photo Composition
This work develops a novel computational means for evaluating the composition aesthetics of a given image based on measuring several well‐grounded composition guidelines and proposes an optimization method for automatically producing a maximally‐aesthetic version of the input image.
Motion Interchange Patterns for Action Recognition in Unconstrained Videos
This paper considers the key elements of motion encoding and focuses on capturing local changes in motion directions, and decouple image edges from motion edges using a suppression mechanism, and compensate for global camera motion by using an especially fitted registration scheme.
Specifying Object Attributes and Relations in Interactive Scene Generation
  • Oron Ashual, L. Wolf
  • Computer Science
    IEEE/CVF International Conference on Computer…
  • 11 September 2019
The method separates between a layout embedding and an appearance embedding, which leads to generated images that better match the scene graph, have higher visual quality, and support more complex scene graphs.
Semi-automatic stereo extraction from video footage
A semi-automatic system that converts conventional video shots to stereoscopic video pairs using a diffusion scheme and a classification scheme that assigns depth to image patches, which tolerates both scene motion and camera motion.
Facial Action Coding
Transformer Interpretability Beyond Attention Visualization
This work proposes a novel way to compute relevancy for Transformer networks that assigns local relevance based on the Deep Taylor Decomposition principle and then propagates these releVancy scores through the layers.
Voice Separation with an Unknown Number of Multiple Speakers
A new method is presented for separating a mixed audio sequence, in which multiple voices speak simultaneously, that greatly outperforms the current state of the art, which, as it is shown, is not competitive for more than two speakers.
Fuzzy Vault