mrhwl123/BFT
注意:源文件中coco格式的json标注有误,使用前需要: sed -i 's/Gr3002/Gr3001/g' annotations_coco/val_v1.5.json BFT (Bird Flocking Tracking) BFT 是面向高动态开放世界目标的多目标跟踪基准,聚焦于严重形变、快速运动与频繁遮挡的场景。数据集由鸟群飞行视频构成,为 CVPR 2024 论文 NetTrack 所提出。 数据规模 106 段精选片段,来自约 6 小时素材 22 个鸟类物种,14 类自然与文化场景,覆盖六大洲 划分:train 35 / val 25 / test 36 视频与标注均为 25 FPS,标注为多目标跟踪包围框 高动态性 相比现有开放世界 MOT 数据集,BFT 具有更分散的宽高比变化(ARC)分布与更大的目标位移,代表更强的动态性,专门用于评估跟踪器在剧烈形变下的鲁棒性。 引用 @inproceedings{nettrack2024cvpr… See the full description on the dataset page: https://huggingface.co/datasets/mrhwl123/BFT.
| Author / organization | mrhwl123 |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 193 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 25 August 2026 |
| Last updated at source | 18 September 2026 |
Popularity, last 2 days
193 now (-6 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
More from mrhwl123
- mrhwl123/MOT17Dataset
- mrhwl123/AnimalTrackDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.