Research map / Detection and segmentation / Tracking
Mode-Conditioned Residual Calibration for Multi-Object Tracking
Muyu Li, Henan Hu, Deepak Jain, Xudong Zhao
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 prototype-based error refinement (ablation) | CAMELTrack · DanceTrack val | HOTA: 66.8 -> 67.6 | Improved |
| add residual vector quantization (ablation) | CAMELTrack + PR · DanceTrack val | HOTA: 67.6 -> 67.9 | Improved |
| add uncertainty-gated denoiser (ablation) | CAMELTrack + PR + VQ · DanceTrack val | HOTA: 67.9 -> 69.0 | Improved |
| add motion cues to the mode classifier (ablation) | MCRC · DanceTrack val | HOTA: 68.0 -> 69.0 | Improved |
| add geometry cues (ablation) | MCRC · DanceTrack val | HOTA: 68.4 -> 69.0 | Improved |
| add confidence cues (ablation) | MCRC · DanceTrack val | HOTA: 68.7 -> 69.0 | Improved |
| use 6 prototype clusters instead of 3 (ablation) | MCRC · DanceTrack val | HOTA: 68.2 -> 69.0 | Improved |
| use 9 or 12 clusters instead of 6 (ablation) | MCRC · DanceTrack val | HOTA: 69.0 -> 69.1 (9 clusters), 69.0 (12 clusters) | No clear effect |
| add magnitude clip to margin test in denoiser acceptance (ablation) | MCRC denoiser · DanceTrack val | HOTA: 68.5 -> 69.0 | Improved |
| use two-level 2x256 codebook instead of 1x64 (ablation) | residual VQ · DanceTrack val | HOTA: 67.4 -> 69.0 | Improved |
| double codebook from 2x128 to 2x256 (ablation) | residual VQ · DanceTrack val | HOTA: 68.9 -> 69.0 | No clear effect |
| align VQ codes to error-mode clusters (ablation) | residual VQ · DanceTrack val | HOTA: 68.7 -> 69.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- ByteTrack — ByteTrack: Multi-Object Tracking by Associating Every Detection Box (ECCV 2022) · better on 4 of 4 shared results
- MeMOTR — MeMOTR: Long-Term Memory-Augmented Transformer for Multi-Object Tracking (ICCV 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- UCMCTrack: Multi-Object Tracking with Uniform Camera Motion Compensation (AAAI 2024 · similarity 0.86)
- 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)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · similarity 0.85)
- Hybrid-SORT: Weak Cues Matter for Online Multi-Object Tracking (AAAI 2024 · similarity 0.85)
- Towards Generalizable Multi-Object Tracking (CVPR 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.
