Research map / Detection and segmentation / Tracking
TubeTK: Adopting Tubes to Track Multi-Object in a One-Step Training Model
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use TubeTK without external detections instead of the strongest listed detection-based baseline | MOT17 online tracking · MOT17 | MOTA: 60.0 -> 63.0 | Helped |
| Use TubeTK without external detections instead of Tracktor | MOT16 online tracking · MOT16 | MOTA: 54.5 -> 64.0 | Helped |
| Use TubeTK with POI detections instead of POI | MOT16 tracking · MOT16 | MOTA: 66.1 -> 66.9 | Helped |
| Use TubeTK without external detections instead of RAN | MOT15 online tracking · MOT15 | MOTA: 56.5 -> 58.4 | Helped |
| Shorten input clips to l=4 (ablation) | TubeTK · MOT17 train | MOTA: 76.9 -> 60.3 | Hurt |
| Fix all Btube lengths to 8 (ablation) | TubeTK · MOT17 train | MOTA: 76.9 -> 74.3 | Hurt |
| Use dynamically selected Btube lengths rather than fixed lengths (ablation) | TubeTK · MOT17 train | FN: 19452 -> 11801 | Helped |
| Replace Tube GIoU loss with Tube IoU loss (ablation) | TubeTK · MOT17 train | MOTA: 76.9 -> 70.5 | Hurt |
| Replace Tube NMS with original NMS (ablation) | TubeTK · MOT17 train | MOTA: 76.9 -> 75.3 | Hurt |
| Use Tube GIoU loss instead of Tube IoU loss (ablation) | TubeTK · MOT17 train | FN: 18148 -> 11801 | Helped |
| Use Tube NMS instead of original NMS (ablation) | TubeTK · MOT17 train | FN: 13421 -> 11801 | Helped |
Closest papers by meaning
- DiffusionTrack: Diffusion Model for Multi-Object Tracking (AAAI 2024 · similarity 0.88)
- OVT-B: A New Large-Scale Benchmark for Open-Vocabulary Multi-Object Tracking (NeurIPS 2024 · similarity 0.88)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · similarity 0.86)
- Multiple Object Tracking as ID Prediction (CVPR 2025 · similarity 0.86)
- SUTrack: Towards Simple and Unified Single Object Tracking (AAAI 2025 · similarity 0.85)
- VastTrack: Vast Category Visual Object Tracking (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.