Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.12832 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915395289481216 |
|---|---|
| author | Kondo, Yuki Ukita, Norimichi Kanayama, Riku Yoshida, Yuki Yamaguchi, Takayuki Yu, Xiang Liang, Guang Liu, Xinyao Wang, Guan-Zhang Chu, Wei-Ta Chuang, Bing-Cheng Lee, Jia-Hua Kuo, Pin-Tseng Chu, I-Hsuan Hsiao, Yi-Shein Wu, Cheng-Han Wu, Po-Yi Tsou, Jui-Chien Liu, Hsuan-Chi Lee, Chun-Yi Yang, Yuan-Fu Shigematsu, Kosuke Shin, Asuka Tran, Ba |
| author_facet | Kondo, Yuki Ukita, Norimichi Kanayama, Riku Yoshida, Yuki Yamaguchi, Takayuki Yu, Xiang Liang, Guang Liu, Xinyao Wang, Guan-Zhang Chu, Wei-Ta Chuang, Bing-Cheng Lee, Jia-Hua Kuo, Pin-Tseng Chu, I-Hsuan Hsiao, Yi-Shein Wu, Cheng-Han Wu, Po-Yi Tsou, Jui-Chien Liu, Hsuan-Chi Lee, Chun-Yi Yang, Yuan-Fu Shigematsu, Kosuke Shin, Asuka Tran, Ba |
| contents | Small Multi-Object Tracking (SMOT) is particularly challenging when targets occupy only a few dozen pixels, rendering detection and appearance-based association unreliable. Building on the success of the MVA2023 SOD4SB challenge, this paper introduces the SMOT4SB challenge, which leverages temporal information to address limitations of single-frame detection. Our three main contributions are: (1) the SMOT4SB dataset, consisting of 211 UAV video sequences with 108,192 annotated frames under diverse real-world conditions, designed to capture motion entanglement where both camera and targets move freely in 3D; (2) SO-HOTA, a novel metric combining Dot Distance with HOTA to mitigate the sensitivity of IoU-based metrics to small displacements; and (3) a competitive MVA2025 challenge with 78 participants and 308 submissions, where the winning method achieved a 5.1x improvement over the baseline. This work lays a foundation for advancing SMOT in UAV scenarios with applications in bird strike avoidance, agriculture, fisheries, and ecological monitoring. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_12832 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MVA 2025 Small Multi-Object Tracking for Spotting Birds Challenge: Dataset, Methods, and Results Kondo, Yuki Ukita, Norimichi Kanayama, Riku Yoshida, Yuki Yamaguchi, Takayuki Yu, Xiang Liang, Guang Liu, Xinyao Wang, Guan-Zhang Chu, Wei-Ta Chuang, Bing-Cheng Lee, Jia-Hua Kuo, Pin-Tseng Chu, I-Hsuan Hsiao, Yi-Shein Wu, Cheng-Han Wu, Po-Yi Tsou, Jui-Chien Liu, Hsuan-Chi Lee, Chun-Yi Yang, Yuan-Fu Shigematsu, Kosuke Shin, Asuka Tran, Ba Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Small Multi-Object Tracking (SMOT) is particularly challenging when targets occupy only a few dozen pixels, rendering detection and appearance-based association unreliable. Building on the success of the MVA2023 SOD4SB challenge, this paper introduces the SMOT4SB challenge, which leverages temporal information to address limitations of single-frame detection. Our three main contributions are: (1) the SMOT4SB dataset, consisting of 211 UAV video sequences with 108,192 annotated frames under diverse real-world conditions, designed to capture motion entanglement where both camera and targets move freely in 3D; (2) SO-HOTA, a novel metric combining Dot Distance with HOTA to mitigate the sensitivity of IoU-based metrics to small displacements; and (3) a competitive MVA2025 challenge with 78 participants and 308 submissions, where the winning method achieved a 5.1x improvement over the baseline. This work lays a foundation for advancing SMOT in UAV scenarios with applications in bird strike avoidance, agriculture, fisheries, and ecological monitoring. |
| title | MVA 2025 Small Multi-Object Tracking for Spotting Birds Challenge: Dataset, Methods, and Results |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2507.12832 |