Motion Focus Recognition in Fast-Moving Egocentric Video
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908944808542208 |
|---|---|
| author | Hong, Si-En Tribble, James Lake, Alexander Wang, Hao Zhou, Chaoyi Bastola, Ashish Huang, Siyu Chaudhary, Eisa Canada, Brian Arslan-Ari, Ismahan Razi, Abolfazl |
| author_facet | Hong, Si-En Tribble, James Lake, Alexander Wang, Hao Zhou, Chaoyi Bastola, Ashish Huang, Siyu Chaudhary, Eisa Canada, Brian Arslan-Ari, Ismahan Razi, Abolfazl |
| contents | From Vision-Language-Action (VLA) systems to robotics, existing egocentric datasets primarily focus on action recognition tasks, while largely overlooking the inherent role of motion analysis in sports and other fast-movement scenarios. To bridge this gap, we propose a real-time motion focus recognition method that estimates the subject's locomotion intention from any egocentric video. We leverage the foundation model for camera pose estimation and introduce system-level optimizations to enable efficient and scalable inference. Evaluated on a collected egocentric action dataset, our method achieves real-time performance with manageable memory consumption through a sliding batch inference strategy. This work makes motion-centric analysis practical for edge deployment and offers a complementary perspective to existing egocentric studies on sports and fast-movement activities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_07154 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Motion Focus Recognition in Fast-Moving Egocentric Video Hong, Si-En Tribble, James Lake, Alexander Wang, Hao Zhou, Chaoyi Bastola, Ashish Huang, Siyu Chaudhary, Eisa Canada, Brian Arslan-Ari, Ismahan Razi, Abolfazl Computer Vision and Pattern Recognition From Vision-Language-Action (VLA) systems to robotics, existing egocentric datasets primarily focus on action recognition tasks, while largely overlooking the inherent role of motion analysis in sports and other fast-movement scenarios. To bridge this gap, we propose a real-time motion focus recognition method that estimates the subject's locomotion intention from any egocentric video. We leverage the foundation model for camera pose estimation and introduce system-level optimizations to enable efficient and scalable inference. Evaluated on a collected egocentric action dataset, our method achieves real-time performance with manageable memory consumption through a sliding batch inference strategy. This work makes motion-centric analysis practical for edge deployment and offers a complementary perspective to existing egocentric studies on sports and fast-movement activities. |
| title | Motion Focus Recognition in Fast-Moving Egocentric Video |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.07154 |