Research map / Detection and segmentation / Tracking
Learn To Match: Automatic Matching Network Design for Visual Tracking
Zhipeng Zhang, Yihao Liu, Xiao Wang, Bing Li, Weiming Hu
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 |
|---|---|---|---|
| replace depthwise cross-correlation with Concatenation | Ocean · OTB100 | success: 67.2 -> 68.4 | Improved |
| replace depthwise cross-correlation with Simple-Transformer | Ocean · OTB100 | success: 67.2 -> 65.8 | Hurt |
| replace depthwise cross-correlation with Transductive-Guidance | Ocean · OTB100 | success: 67.2 -> 65.0 | Hurt |
| replace depthwise cross-correlation with Pairwise-Relation | Ocean · OTB100 | success: 67.2 -> 67.8 | Improved |
| replace depthwise cross-correlation with Pointwise-Addition | Ocean · OTB100 | success: 67.2 -> 67.1 | No clear effect |
| add searched matching networks (BCM) to the tracker | Ocean · OTB100 | success: 67.2 -> 71.4 | Improved |
| add searched matching networks (BCM) to the tracker | Ocean · LaSOT | success: 52.6 -> 58.3 | Improved |
| add searched matching networks (BCM) to the tracker | Ocean · TrackingNet | success: 70.3 -> 76.0 | Improved |
| add searched matching networks (BCM) to the tracker | Ocean · GOT10K | AO: 59.2 -> 65.2 | Improved |
| add per-channel manipulators instead of one scalar per operator (ablation) | AutoMatch · OTB100 | success: 69.5 -> 71.4 | Improved |
| add per-channel manipulators instead of one scalar per operator (ablation) | AutoMatch · LaSOT | success: 54.7 -> 58.3 | Improved |
| use BCM search instead of random operator selection (ablation) | AutoMatch · OTB100 | success: 69.1 -> 71.4 | Improved |
Closest papers by meaning
- SiamRPN++: Evolution of Siamese Visual Tracking With Very Deep Networks (CVPR 2019 · similarity 0.89)
- Learning Dynamic Siamese Network for Visual Object Tracking (ICCV 2017 · similarity 0.89)
- SiamFC++: Towards Robust and Accurate Visual Tracking with Target Estimation Guidelines (AAAI 2020 · similarity 0.89)
- Joint Classification and Regression for Visual Tracking with Fully Convolutional Siamese Networks (IJCV 2022 · similarity 0.88)
- GradNet: Gradient-Guided Network for Visual Object Tracking (ICCV 2019 · 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.
