Research map / Detection and segmentation / Tracking
Learning Dynamic Siamese Network for Visual Object Tracking
Qing Guo, Wei Feng, Ce Zhou, Rui Huang, Liang Wan, Song Wang
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 online target-variation and background-suppression transformations to static SiamFC (DSiam, conv5 only) | SiamFC · OTB-2013 | AUC of success plot: 5.6% relative improvement over SiamFC | Improved |
| add elementwise multi-layer fusion of conv4 and conv5 (DSiamM vs DSiam) | DSiam · OTB-2013 | AUC of success plot: 5.6% -> 7.9% relative improvement over SiamFC | Improved |
| add elementwise multi-layer fusion (DSiamM vs DSiam) | DSiam · VOT-2015 | overlap: 0.2804 -> 0.2927 | Improved |
| add elementwise multi-layer fusion (DSiamM vs DSiam) | DSiam · VOT-2015 | accuracy: 0.5414 -> 0.5566 | Improved |
| add the target appearance variation transformation (VT) (ablation) | DSiamM · OTB-2013 | tracking accuracy | Improved |
| add the background suppression transformation (BS) (ablation) | DSiamM · OTB-2013 | tracking accuracy | Improved |
| add elementwise multi-layer fusion (Multi) (ablation) | DSiamM · OTB-2013 | tracking accuracy | Improved |
| learn variation transformation instead of simply updating the template O1 to Ot-1 (ablation) | DSiamM · OTB-2013 | tracking accuracy | Improved |
| use offline-learned elementwise fusion weights instead of fixed fusion weights (ablation) | DSiamM · OTB-2013 | tracking accuracy | Improved |
| use DSiamM instead of the retrained MDNet | R-MDNet · OTB-2013 | AUC of success plot: 6% relative improvement | Improved |
| use DSiamM instead of HCF | HCF · OTB-2013 | AUC of success plot: 8.4% relative improvement | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Staple — Staple: Complementary Learners for Real-Time Tracking (CVPR 2016) · better on 2 of 3 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
- Discriminative and Robust Online Learning for Siamese Visual Tracking (AAAI 2020 · similarity 0.92)
- Deeper and Wider Siamese Networks for Real-Time Visual Tracking (CVPR 2019 · similarity 0.91)
- Joint Classification and Regression for Visual Tracking with Fully Convolutional Siamese Networks (IJCV 2022 · similarity 0.91)
- Visual Object Tracking with Discriminative Filters and Siamese Networks: A Survey and Outlook (TPAMI 2022 · similarity 0.91)
- High Performance Visual Tracking With Siamese Region Proposal Network (CVPR 2018 · similarity 0.91)
- Siamese Box Adaptive Network for Visual Tracking (CVPR 2020 · similarity 0.91)
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.
