Research map / Detection and segmentation / Tracking
Tracking Through Containers and Occluders in the Wild
Read the paperCheck 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 |
|---|---|---|---|
| Predict target, occluder, and container masks with the TCOW TimeSFormer model rather than AOT | Kubric Random and Kubric Containers · Kubric synthetic benchmarks | target IoU (all): 41.3 -> 53.0; 26.5 -> 36.8 | Helped |
| Predict the target and its occluder/container explicitly rather than using AOT | TCOW · Kubric Random and Kubric Containers | occluder IoU: 5.1 -> 70.5; 6.8 -> 76.8 | Helped |
| Train with X-ray annotations instead of visible pixels only (ablation) | TCOW · Kubric Random | target IoU (invisible): 0.1 -> 16.6 | Helped |
| Train with X-ray annotations instead of visible pixels only (ablation) | TCOW · Kubric Containers | target IoU (invisible): 0.1 -> 16.0 | Helped |
| Train on realistic textured scenes rather than flat-color cartoon data (ablation) | TCOW · Kubric Random | occluder IoU: 30.0 -> 70.5 | Helped |
| Train on realistic textured scenes rather than flat-color cartoon data (ablation) | TCOW · Kubric Containers | container IoU: 40.1 -> 78.2 | Helped |
| Use visible-only training rather than full X-ray supervision (ablation) | TCOW · Rubric Office | occluder IoU: 30.1 -> 39.2 | Helped |
| Use visible-only training rather than full X-ray supervision (ablation) | TCOW · Rubric Cup Games | occluder IoU: 35.0 -> 27.6 | Hurt |
| Train on flat-color cartoon data rather than realistic synthetic scenes (ablation) | TCOW · Rubric Cup Games | container IoU: 7.6 -> 14.3 | Helped |
| Train on flat-color cartoon data rather than realistic synthetic scenes (ablation) | TCOW · Kubric Random | target IoU (all): 53.0 -> 31.3 | Hurt |
| Use the proposed TCOW model rather than AOT | Rubric Office · Rubric Office | target IoU: 78.2 -> 69.4 | Hurt |
| Use the proposed TCOW model rather than AOT | Rubric Cup Games · Rubric Cup Games | target IoU: 50.2 -> 38.3 | Hurt |
Closest papers by meaning
- Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision (NeurIPS 2025 · similarity 0.89)
- Track-On: Transformer-based Online Point Tracking with Memory (ICLR 2025 · similarity 0.88)
- A Distractor-Aware Memory for Visual Object Tracking with SAM2 (CVPR 2025 · similarity 0.88)
- TGFormer: Transformer with Track Query Group for Multi-Object Tracking (AAAI 2025 · similarity 0.87)
- ODTrack: Online Dense Temporal Token Learning for Visual Tracking (AAAI 2024 · similarity 0.87)
- TrackIME: Enhanced Video Point Tracking via Instance Motion Estimation (NeurIPS 2024 · similarity 0.87)
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.