Research map / Detection and segmentation / Tracking
Unified Transformer Tracker for Object Tracking
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 UTT instead of UniTrack | Unified tracking on LaSOT · LaSOT test | Success: 35.1 -> 64.6 | Helped |
| Use UTT instead of TransT | SOT tracking · LaSOT test | Success: 64.9 -> 64.6 | No clear effect |
| Use UTT instead of TransT | SOT tracking · TrackingNet test | Success: 81.4 -> 79.7 | Hurt |
| Use UTT instead of TransT | SOT tracking · GOT-10k test | AO: 72.3 -> 67.2 | Hurt |
| Use UTT-FairMOT instead of FairMOT | MOT16 tracking with FairMOT detections · MOT16 test | MOTP: 80.6 -> 81.4 | Helped |
| Add the proposal decoder (ablation) | UTT · LaSOT test | Success: 48.1 -> 58.6 | Helped |
| Add the target decoder to the proposal-decoder model (ablation) | UTT · LaSOT test | Success: 58.6 -> 59.2 | Helped |
| Add the target decoder without the proposal decoder (ablation) | UTT · LaSOT test | Success: 48.1 -> 49.8 | Helped |
| Replace MCA with correlation attention and omit MSA (ablation) | UTT target transformer · LaSOT test | Success: 57.1 -> 57.6 | Helped |
| Add MSA to the CorrAtt design (ablation) | UTT target transformer · LaSOT test | Success: 57.6 -> 58.9 | Helped |
| Replace MCA with CorrAtt and omit MSA (ablation) | UTT target transformer · MOT16 half training set | MOTA: 64.4 -> 64.9 | Helped |
| Add MSA to the CorrAtt design (ablation) | UTT target transformer · MOT16 half training set | MOTA: 64.9 -> 65.4 | Helped |
Closest papers by meaning
- SUTrack: Towards Simple and Unified Single Object Tracking (AAAI 2025 · similarity 0.88)
- Revisiting motion information for RGB-Event tracking with MOT philosophy (NeurIPS 2024 · similarity 0.87)
- M3SOT: Multi-Frame, Multi-Field, Multi-Space 3D Single Object Tracking (AAAI 2024 · similarity 0.87)
- UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking (CVPR 2026 · similarity 0.87)
- Is Multiple Object Tracking a Matter of Specialization? (NeurIPS 2024 · similarity 0.87)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · 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.