| Presentation | 2024-01-26 Body Parts Motion Guided Deformable Attention for Action Recognition Yuji Sato, Umihiro Kamoto, Takeo Ueta, Yasunori Ishii, Takayoshi Yamashita, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In recent years, video transformers for action recognition have beenproposed and achieve high performance. Most of the existing video transformerscalculate the attention with pre-designed patches at fixed positions. As a result, patches are compared between specific positions within and between frames, and motion is not taken into account. A dynamic attention-based method has also been proposed that predictsthe region to which the query patch should direct attention based onthe motion between video frames, but it is the motion of theentire video and does not take into account the motion related tothe action class. To address these issues, we propose a video transformer with a deformable attention mechanism based on the motion ofhuman body parts that contribute to human action recognition. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Action Recognition / Video Transformers / Deformable Attention / Body Parts Motion |
| Paper # | PRMU2023-44 |
| Date of Issue | 2024-01-18 (PRMU) |
| Conference Information | |
| Committee | PRMU / MVE / VRSJ-SIG-MR / IPSJ-CVIM |
|---|---|
| Conference Date | 2024/1/25(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Keio Univ. (Hiyoshi Campus) |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Kunio Kashio(NTT) / Kiyoshi Kiyokawa(NAIST) / / 日浦 慎作(兵庫県立大) |
| Vice Chair | Takuya Funatomi(NAIST) / Go Irie(Tokyo Univ. of Science) / Sumaru Niida(KDDI Research) |
| Secretary | Takuya Funatomi(Tokyo Inst. of Tech.) / Go Irie(Riken) / Sumaru Niida(Otsuma Women's University) / (DNP) / (Kyushu Univ.) |
| Assistant | Kei Shimonishi(Kyoto Univ.) / Kensho Hara(AIST) / Hidehiko Shishido(Soka University) / Atsushi Nakazawa(Kyoto Univ.) / Naoya Tojo(KDDI Research) / Naoki Hagiyama(NTT) / Yuji Tatada(Univ. of Tokyo) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Media Experience and Virtual Environment / SIG-MR / Special Interest Group on Computer Vision and Image Media |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Body Parts Motion Guided Deformable Attention for Action Recognition |
| Sub Title (in English) | |
| Keyword(1) | Action Recognition |
| Keyword(2) | Video Transformers |
| Keyword(3) | Deformable Attention |
| Keyword(4) | Body Parts Motion |
| Keyword(5) | |
| 1st Author's Name | Yuji Sato |
| 1st Author's Affiliation | Panasonic Connect Co., Ltd.(Panasonic Connect) |
| 2nd Author's Name | Umihiro Kamoto |
| 2nd Author's Affiliation | Panasonic Connect Co., Ltd.(Panasonic Connect) |
| 3rd Author's Name | Takeo Ueta |
| 3rd Author's Affiliation | Panasonic Connect Co., Ltd.(Panasonic Connect) |
| 4th Author's Name | Yasunori Ishii |
| 4th Author's Affiliation | Panasonic Holdings Corporation(Panasonic Holdings) |
| 5th Author's Name | Takayoshi Yamashita |
| 5th Author's Affiliation | Chubu University(Chubu Univ.) |
| Date | 2024-01-26 |
| Paper # | PRMU2023-44 |
| Volume (vol) | vol.123 |
| Number (no) | PRMU-358 |
| Page | pp.pp.26-31(PRMU), |
| #Pages | 6 |
| Date of Issue | 2024-01-18 (PRMU) |