Chokri Ben Amar

Papers

Total Citations

H-Index

About

Chokri Ben Amar is a computer vision researcher whose work focuses on human action recognition, scene understanding, and robotic perception. His most influential contribution, the "SIFT Accordion: A Space-Time Descriptor Applied to Human Action Recognition" (2011, 5 citations), introduced a novel spatiotemporal descriptor that captures both appearance and motion information for recognizing human activities in video. This work addresses key challenges in video surveillance, human-machine interaction, and sports video retrieval by encoding local features across space and time. Ben Amar also led the REGIM team's participation in the ImageCLEF 2013 Robot Vision Challenge, developing methods for objects and scenes classification in indoor environments—a critical capability for autonomous robot navigation. Additionally, his "Skyline-based approach for natural scene identification" (2016, 3 citations) leverages geometric skyline features for geo-localization and aerial robotics, demonstrating his versatility in applying computer vision to real-world robotics problems. While his citation counts reflect focused contributions in specialized areas, Ben Amar's work bridges fundamental computer vision techniques with practical applications in robotics and surveillance, making him a notable figure in the REGIM research community.