Research map / Detection and segmentation / Tracking
Chained-Tracker: Chaining Paired Attentive Regression Results for End-to-End Joint Multiple-Object Detection and Tracking
Jinlong Peng, Changan Wang, Fangbin Wan, Yang Wu, Yabiao Wang, Ying Tai, Chengjie Wang, Jilin Li, Feiyue Huang, Yanwei Fu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add object attention (classification confidence map) to the paired regression branch (ablation) | CTracker baseline · MOT17 test | MOTA: 64.4 -> 66.0 | Improved |
| add object attention (classification confidence map) to the paired regression branch (ablation) | CTracker baseline · MOT17 test | IDF1: 51.6 -> 55.7 | Improved |
| add object attention to the baseline (ablation) | CTracker baseline · MOT17 test | MOTP: 78.2 -> 78.8 | Improved |
| add a stand-alone ID verification branch without using it as attention (ablation) | Baseline+ObjAtten · MOT17 test | MOTA: 66.0 -> 65.6 | Hurt |
| add a stand-alone ID verification branch without using it as attention (ablation) | Baseline+ObjAtten · MOT17 test | IDF1: 55.7 -> 55.2 | Hurt |
| use ID verification output as additional attention (joint attention) (ablation) | Baseline+ObjAtten · MOT17 test | MOTA: 66.0 -> 66.6 | Improved |
| use ID verification output as additional attention (joint attention) (ablation) | Baseline+ObjAtten · MOT17 test | IDF1: 55.7 -> 57.4 | Improved |
| use ID verification output as additional attention (joint attention) (ablation) | Baseline+ObjAtten · MOT17 test | MOTP: 78.8 -> 78.2 | Hurt |
| add Memory Sharing Mechanism (feature reuse across chain nodes) (ablation) | CTracker · Inference time, 1080x1920 | Total time (ms): 223.56 -> 148.10 | Improved |
| use CTracker detections instead of public detections for Tracktor | Tracktor · MOT17 test | MOTA: 53.5 -> 54.4 | Improved |
| use CTracker detections instead of public detections for Tracktor | Tracktor · MOT17 test | IDF1: 52.3 -> 56.1 | Improved |
| replace Tracktor+CTdet tracking with CTracker's chained paired regression | Tracktor+CTdet · MOT17 test (private detection) | MOTA: 54.4 -> 66.6 | Improved |
Closest papers by meaning
- Multiple Object Tracking as ID Prediction (CVPR 2025 · similarity 0.88)
- COVTrack++: Learning Open-Vocabulary Multi-Object Tracking from Continuous Videos via a Synergistic Paradigm (TPAMI 2026 · similarity 0.88)
- Tracking Without Bells and Whistles (ICCV 2019 · similarity 0.87)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · similarity 0.87)
- DiffusionTrack: Diffusion Model for Multi-Object Tracking (AAAI 2024 · similarity 0.86)
- Self-Supervised Multi-Object Tracking with Path Consistency (CVPR 2024 · similarity 0.86)
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.
