Research map / Detection and segmentation / Tracking
CLNet: A Compact Latent Network for Fast Adjusting Siamese Trackers
Xingping Dong, Jianbing Shen, Ling Shao, Fatih Porikli
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 |
|---|---|---|---|
| retrain SiamRPN++ with the new training method only (ablation) | SiamRPN++ · Combined DTB70+NFS30+UAV123 | Precision (%): 73.3 -> 74.3 | Mixed |
| retrain SiamRPN++ with the new training method only (ablation) | SiamRPN++ · Combined DTB70+NFS30+UAV123 | AUC (%): 57.6 -> 57.4 | Hurt |
| add adjustment network without latent encoder (ablation) | SiamRPN++ · Combined DTB70+NFS30+UAV123 | Precision (%): 73.3 -> 73.8 | Improved |
| add latent encoder (ablation) | adjustment network · Combined DTB70+NFS30+UAV123 | Precision (%): 73.8 -> 75.5 | Improved |
| add latent encoder (ablation) | adjustment network · Combined DTB70+NFS30+UAV123 | AUC (%): 58.2 -> 60.1 | Improved |
| sample image pairs from one sequence per training batch (ablation) | CLNet · Combined DTB70+NFS30+UAV123 | Precision (%): 75.5 -> 76.6 | Improved |
| sample image pairs from one sequence per training batch (ablation) | CLNet · Combined DTB70+NFS30+UAV123 | AUC (%): 60.1 -> 60.2 | No clear effect |
| add diverse sample mining (ablation) | CLNet · Combined DTB70+NFS30+UAV123 | Precision (%): 76.6 -> 77.2 | Improved |
| add diverse sample mining (ablation) | CLNet · Combined DTB70+NFS30+UAV123 | AUC (%): 60.2 -> 61.1 | Improved |
| replace SiamRPN++ with CLNet | SiamRPN++ · UAV123 | AUC: 0.611 -> 0.633 | Improved |
| replace SiamRPN++ with CLNet | SiamRPN++ · LaSOT | AUC (%): 49.3 -> 49.9 | Improved |
Later papers that compared with this method
1 later accepted paper reports CLNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Object Tracking by Jointly Exploiting Frame and Event Domain (ICCV 2021 · better on 2 of 2 shared results)
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
- Adaptive Siamese Tracking with a Compact Latent Network (TPAMI 2022 · similarity 0.96)
- SiamRPN++: Evolution of Siamese Visual Tracking With Very Deep Networks (CVPR 2019 · similarity 0.89)
- Deeper and Wider Siamese Networks for Real-Time Visual Tracking (CVPR 2019 · similarity 0.89)
- Learning the Model Update for Siamese Trackers (ICCV 2019 · similarity 0.89)
- Discriminative and Robust Online Learning for Siamese Visual Tracking (AAAI 2020 · similarity 0.88)
- Learning Random Noise Salient Feature Fusion Siamese Network for Low-Resolution Object Tracking (Student Abstract) (AAAI 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.
