{"ID":2895587,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.08344","arxiv_id":"2507.08344","title":"MM-Gesture: Towards Precise Micro-Gesture Recognition through Multimodal Fusion","abstract":"In this paper, we present MM-Gesture, the solution developed by our team HFUT-VUT, which ranked 1st in the micro-gesture classification track of the 3rd MiGA Challenge at IJCAI 2025, achieving superior performance compared to previous state-of-the-art methods. MM-Gesture is a multimodal fusion framework designed specifically for recognizing subtle and short-duration micro-gestures (MGs), integrating complementary cues from joint, limb, RGB video, Taylor-series video, optical-flow video, and depth video modalities. Utilizing PoseConv3D and Video Swin Transformer architectures with a novel modality-weighted ensemble strategy, our method further enhances RGB modality performance through transfer learning pre-trained on the larger MA-52 dataset. Extensive experiments on the iMiGUE benchmark, including ablation studies across different modalities, validate the effectiveness of our proposed approach, achieving a top-1 accuracy of 73.213%. Code is available at: https://github.com/momiji-bit/MM-Gesture.","short_abstract":"In this paper, we present MM-Gesture, the solution developed by our team HFUT-VUT, which ranked 1st in the micro-gesture classification track of the 3rd MiGA Challenge at IJCAI 2025, achieving superior performance compared to previous state-of-the-art methods. MM-Gesture is a multimodal fusion framework designed specif...","url_abs":"https://arxiv.org/abs/2507.08344","url_pdf":"https://arxiv.org/pdf/2507.08344v2","authors":"[\"Jihao Gu\",\"Fei Wang\",\"Kun Li\",\"Yanyan Wei\",\"Zhiliang Wu\",\"Dan Guo\"]","published":"2025-07-11T06:45:42Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Transformer\"]","has_code":false,"code_links":[{"ID":612206,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2895587,"paper_url":"https://arxiv.org/abs/2507.08344","paper_title":"MM-Gesture: Towards Precise Micro-Gesture Recognition through Multimodal Fusion","repo_url":"https://github.com/momiji-bit/MM-Gesture","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}