Research map / Detection and segmentation / Tracking
Progressive Unsupervised Learning for Visual 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 |
|---|---|---|---|
| add anchor-based hard negative mining (ablation) | BD learning in AlexPUL · OTB-13 | AUC: 0.491 -> 0.515 | Helped |
| add anchor-based hard negative mining (ablation) | BD learning in ResPUL · OTB-13 | AUC: 0.580 -> 0.598 | Helped |
| add temporal correspondence learning with 10-frame mining (ablation) | AlexPUL with BD and AHM · OTB-13 | AUC: 0.515 -> 0.567 | Helped |
| increase temporal mining length to 20 frames (ablation) | AlexPUL with temporal correspondence learning · OTB-13 | AUC: 0.567 -> 0.574 | Helped |
| add temporal correspondence learning with 20-frame mining (ablation) | ResPUL with BD and AHM · OTB-13 | AUC: 0.598 -> 0.620 | Helped |
| use NR loss instead of BCE for noisy temporal patch pairs (ablation) | AlexPUL with TCL-10 · OTB-13 | AUC: 0.530 -> 0.567 | Helped |
| use NR loss instead of BCE for noisy temporal patch pairs (ablation) | AlexPUL with TCL-20 · OTB-13 | AUC: 0.526 -> 0.574 | Helped |
| train the unsupervised tracker without annotated videos or online adaptation | ResPUL · VOT16 | accuracy: 0.530 -> 0.554 | Mixed |
| train the unsupervised tracker without annotated videos or online adaptation | ResPUL · VOT16 | EAO: 0.303 -> 0.263 | Hurt |
| train the unsupervised tracker without annotated videos or online adaptation | ResPUL · TrackingNet | Success: 0.543 -> 0.546 | Helped |
Closest papers by meaning
- Unsupervised Learning of Accurate Siamese Tracking (CVPR 2022 · similarity 0.91)
- Unsupervised Deep Tracking (CVPR 2019 · similarity 0.91)
- Learning a Visual Tracker from a Single Movie without Annotation (AAAI 2019 · similarity 0.89)
- Discriminative and Robust Online Learning for Siamese Visual Tracking (AAAI 2020 · similarity 0.88)
- Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision (NeurIPS 2025 · similarity 0.85)
- Learning Random Noise Salient Feature Fusion Siamese Network for Low-Resolution Object Tracking (Student Abstract) (AAAI 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.