Visual Object Tracking On Didi
評価指標
Tracking quality
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Tracking quality | Paper Title | Repository |
|---|---|---|---|
| Cutie | 0.575 | Putting the Object Back into Video Object Segmentation | |
| SAMURAI | 0.680 | SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory | |
| ODTrack | 0.608 | ODTrack: Online Dense Temporal Token Learning for Visual Tracking | |
| DAM4SAM | 0.694 | A Distractor-Aware Memory for Visual Object Tracking with SAM2 | |
| KeepTrack | 0.502 | Learning Target Candidate Association to Keep Track of What Not to Track | |
| AQATrack | 0.535 | Autoregressive Queries for Adaptive Tracking with Spatio-TemporalTransformers | - |
| SeqTrack | 0.529 | - | - |
| SAM2.1 | 0.649 | SAM 2: Segment Anything in Images and Videos | |
| AOT | 0.541 | AOT: Appearance Optimal Transport Based Identity Swapping for Forgery Detection | - |
| TransT | 0.465 | Transformer Tracking | |
| SAM2.1Long | 0.646 | SAM2Long: Enhancing SAM 2 for Long Video Segmentation with a Training-Free Memory Tree |
0 of 11 row(s) selected.