Research map / Label-efficient and robust learning / Recognition and long-tail
Ranking-Based Siamese Visual 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 classification ranking loss (ablation) | SiamRPN++ and SiamBAN · Combined TC128, NFS30, UAV123 | AUC: 57.93 -> 60.63; 59.61 -> 61.15 | Helped |
| Add the proposed IoU-guided ranking loss to classification ranking (ablation) | SiamRPN++ and SiamBAN · Combined TC128, NFS30, UAV123 | AUC: 60.63 -> 62.08; 61.15 -> 62.24 | Helped |
| Use the original IoU-guided ranking loss rather than the modified IGR (ablation) | SiamRPN++ and SiamBAN with CR · Combined TC128, NFS30, UAV123 | AUC: 60.63 -> 60.87 vs 62.08; 61.15 -> 60.92 vs 62.24 | Hurt |
| Replace hard-negative mining with cross-entropy plus classification ranking loss (ablation) | Siamese tracker classification · Confidence-map analysis | distractor suppression | Helped |
| Add RBO to SiamRPN++ | SiamRPN++ · TC128 | AUC: 57.3 -> 61.9 | Helped |
| Add RBO to Siamese trackers | SiamRPN++ and SiamBAN · VOT2016 | EAO: absolute gain of 4.9% | Helped |
| Add RBO to Siamese trackers | SiamRPN++ and SiamBAN · LaSOT test | AUC: 3.6%; 1.1% | Helped |
| Add RBO to SiamRPN++ | SiamRPN++ · GOT-10k test | AO: 8.5% | Helped |
| Add RBO to TransT | TransT · LaSOT, GOT-10k, UAV123 | benchmark performance: 64.9 -> 65.6 AUC; 72.3 -> 72.7 AO; 67.9 -> 68.5 AUC | Helped |
| Replace depth-wise correlation with pixel-wise correlation (ablation) | SiamBAN-RBO · GOT-10k test | AO: 60.8 -> 64.4 | Helped |
| Use RBO with Siamese trackers | SiamRPN++-RBO, SiamBAN-RBO, SiamPW-RBO · OTB100 | AUC: 69.6 -> 69.9; 69.6 -> 70.1; 69.8 | Mixed |
Closest papers by meaning
- SiamFC++: Towards Robust and Accurate Visual Tracking with Target Estimation Guidelines (AAAI 2020 · similarity 0.91)
- Graph Attention Tracking (CVPR 2021 · similarity 0.90)
- SiamRPN++: Evolution of Siamese Visual Tracking With Very Deep Networks (CVPR 2019 · similarity 0.90)
- High Performance Visual Tracking With Siamese Region Proposal Network (CVPR 2018 · similarity 0.90)
- Deformable Siamese Attention Networks for Visual Object Tracking (CVPR 2020 · similarity 0.90)
- HIPTrack: Visual Tracking with Historical Prompts (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.