Research map / Detection and segmentation / Tracking
Adaptive Siamese Tracking with a Compact Latent Network
Xingping Dong, Jianbing Shen, Fatih Porikli, Jiebo Luo, Ling Shao
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 adjustment network (without latent encoder) to SiamRPN++ (ablation) | SiamRPN++ · NfS30+DTB70 combined | AUC (%): 55.8 -> 57.0 | Improved |
| add latent encoder (mean and std statistics) on top of adjustment network (ablation) | CLNet-RPN · NfS30+DTB70 combined | AUC (%): 57.0 -> 58.9 | Improved |
| sample image pairs from one sequence per batch (ablation) | CLNet-RPN · NfS30+DTB70 combined | precision (%): 72.5 -> 73.8 | Improved |
| add diverse sample mining (ablation) | CLNet-RPN · NfS30+DTB70 combined | AUC (%): 59.1 -> 60.0 | Improved |
| add conditional updating (ablation) | CLNet-RPN · NfS30+DTB70 combined | AUC (%): 60.0 -> 61.0 | Improved |
| add CLNet adjustment to the regression branch as well as classification (ablation) | CLNet-RPN · NfS30+DTB70 combined | AUC (%): 59.0 -> 60.0 | Improved |
| retrain base tracker with the extra training iterations only (ablation) | SiamRPN++ · NfS30+DTB70 combined | AUC (%): 55.8 -> 56.1 | No clear effect |
| use additive weight augmentation instead of FILM (ablation) | CLNet-RPN · NfS30+DTB70 combined | AUC (%): 58.7 -> 60.0 | Improved |
| use additive weight augmentation instead of CBAM (ablation) | CLNet-RPN · NfS30+DTB70 combined | precision (%): 72.2 -> 74.4 | Improved |
| apply CLNet*-RPN adjustment to SiamRPN++ | SiamRPN++ · VOT2019 real-time | EAO: 0.285 -> 0.313 | Improved |
| apply CLNet adjustment to SiamBAN | SiamBAN · VOT2019 real-time | EAO: 0.322 -> 0.356 | Improved |
| apply CLNet adjustment to SiamFC-Pt | SiamFC-Pt · LaSOT test | AUC (%): 33.2 -> 33.3 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- SiamRCNN — Siam R-CNN: Visual Tracking by Re-Detection (CVPR 2020) · better on 0 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
- CLNet: A Compact Latent Network for Fast Adjusting Siamese Trackers (ECCV 2020 · similarity 0.96)
- SiamRPN++: Evolution of Siamese Visual Tracking With Very Deep Networks (CVPR 2019 · similarity 0.91)
- Discriminative and Robust Online Learning for Siamese Visual Tracking (AAAI 2020 · similarity 0.91)
- Learning the Model Update for Siamese Trackers (ICCV 2019 · similarity 0.91)
- Deeper and Wider Siamese Networks for Real-Time Visual Tracking (CVPR 2019 · similarity 0.90)
- Learning Dynamic Siamese Network for Visual Object Tracking (ICCV 2017 · similarity 0.90)
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.
