Research map / Vision-language and multimodal / Cross-modal learning
AlignTrack: Top-Down Spatiotemporal Resolution Alignment for RGB-Event Visual Tracking
Chuanyu Sun, Jiqing Zhang, Yang Wang, Yuanchen Wang, Yutong Jiang, Baocai Yin, Xin Yang
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 TDA module (ablation) | AlignTrack · FE240/CRSOT/VisEvent | SR: -2.7 average SR drop without it | Improved |
| add CPA module instead of learnable tokens (ablation) | AlignTrack · FE240/CRSOT/VisEvent | SR: 0.9% SR decrease without it | Improved |
| add cross-fusion in CPA (ablation) | AlignTrack · FE240 | SR: 67.3 -> 68.1 | Improved |
| add reconstruction loss (ablation) | AlignTrack · FE240 | SR: 66.2 -> 68.1 | Improved |
| add CSA loss (ablation) | AlignTrack · FE240/CRSOT/VisEvent | SR: 1.3% average SR drop without it | Improved |
| add target cropping to CSA loss (ablation) | AlignTrack · FE240 | SR: 65.6 -> 68.1 | Improved |
| use both RGB and Event inputs rather than RGB only (ablation) | AlignTrack · FE240 | SR: 44.2 -> 68.1 | Improved |
| use both modalities rather than Event only (ablation) | AlignTrack · CRSOT | SR: 11.2 -> 63.0 | Improved |
| increase decoder layers from 3 to 12 (ablation) | AlignTrack · FE240 | SR: 63.0 -> 68.1 | Improved |
| inject top-down signal into Value rather than Query (ablation) | AlignTrack · FE240 | SR: 65.7 -> 68.1 | Improved |
| inject top-down signal into Value rather than Key (ablation) | AlignTrack · FE240 | SR: 66.3 -> 68.1 | Improved |
| set lambda_rec to 0.1 instead of 1 (ablation) | AlignTrack · FE240 | SR: 67.9 -> 68.1 | Improved |
Closest papers by meaning
- Temporal Adaptive RGBT Tracking with Modality Prompt (AAAI 2024 · similarity 0.88)
- CSTrack: Enhancing RGB-X Tracking via Compact Spatiotemporal Features (ICML 2025 · similarity 0.88)
- Single-Model and Any-Modality for Video Object Tracking (CVPR 2024 · similarity 0.88)
- Cross-modulated Attention Transformer for RGBT Tracking (AAAI 2025 · similarity 0.87)
- Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline (CVPR 2024 · similarity 0.87)
- UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking (CVPR 2026 · similarity 0.86)
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.
