{"ID":2864500,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.23009","arxiv_id":"2509.23009","title":"Disentangling Static and Dynamic Information for Reducing Static Bias in Action Recognition","abstract":"Action recognition models rely excessively on static cues rather than dynamic human motion, which is known as static bias. This bias leads to poor performance in real-world applications and zero-shot action recognition. In this paper, we propose a method to reduce static bias by separating temporal dynamic information from static scene information. Our approach uses a statistical independence loss between biased and unbiased streams, combined with a scene prediction loss. Our experiments demonstrate that this method effectively reduces static bias and confirm the importance of scene prediction loss.","short_abstract":"Action recognition models rely excessively on static cues rather than dynamic human motion, which is known as static bias. This bias leads to poor performance in real-world applications and zero-shot action recognition. In this paper, we propose a method to reduce static bias by separating temporal dynamic information...","url_abs":"https://arxiv.org/abs/2509.23009","url_pdf":"https://arxiv.org/pdf/2509.23009v1","authors":"[\"Masato Kobayashi\",\"Ning Ding\",\"Toru Tamaki\"]","published":"2025-09-27T00:03:41Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}