Research map / Video / Video segmentation and tracking
End-to-End Flow Correlation Tracking With Spatial-Temporal Attention
Read the paperCheck 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 flow information and aggregate warped frame features (ablation) | FlowTrack · OTB2013 | AUC: 0.625 -> 0.689 | Helped |
| train FlowNet end-to-end rather than fixing it (ablation) | FlowTrack · VOT2015 | EAO: 0.2542 -> 0.3405 | Helped |
| use learned flow aggregation instead of time-decayed fusion (ablation) | FlowTrack · VOT2015 | EAO: 0.2584 -> 0.3405 | Helped |
| add temporal attention to spatial attention (ablation) | FlowTrack · OTB2015 | AUC: 0.642 -> 0.655 | Helped |
| add temporal attention to spatial attention (ablation) | FlowTrack · VOT2016 | EAO: 0.2712 -> 0.3342 | Helped |
| add flow information and aggregate warped frame features (ablation) | FlowTrack · VOT2015 | EAO: 0.2637 -> 0.3405 | Helped |
| add flow information and aggregate warped frame features (ablation) | FlowTrack · VOT2016 | EAO: 0.2404 -> 0.3342 | Helped |
| train FlowNet end-to-end rather than fixing it (ablation) | FlowTrack · VOT2016 | EAO: 0.2291 -> 0.3342 | Helped |
| use learned flow aggregation instead of time-decayed fusion (ablation) | FlowTrack · VOT2016 | EAO: 0.2516 -> 0.3342 | Helped |
| add temporal attention to spatial attention (ablation) | FlowTrack · OTB2013 | AUC: 0.667 -> 0.689 | Helped |
| use end-to-end learned flow instead of fixed FlowNet (ablation) | FlowTrack · OTB2013 | AUC: 0.617 -> 0.689 | Mixed |
| use end-to-end learned flow instead of fixed FlowNet (ablation) | FlowTrack · OTB2015 | AUC: 0.583 -> 0.655 | Helped |
Closest papers by meaning
- FlowTrack: Revisiting Optical Flow for Long-Range Dense Tracking (CVPR 2024 · similarity 0.88)
- Optical Flow in Deep Visual Tracking (AAAI 2020 · similarity 0.88)
- Correlation-Aware Deep Tracking (CVPR 2022 · similarity 0.87)
- Multi-Cue Correlation Filters for Robust Visual Tracking (CVPR 2018 · similarity 0.87)
- Temporal Coherent Object Flow for Multi-Object Tracking (AAAI 2025 · similarity 0.87)
- EDCFlow: Exploring Temporally Dense Difference Maps for Event-based Optical Flow Estimation (CVPR 2025 · similarity 0.85)
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.