Research map / Detection and segmentation / Tracking
HTTrack: Learning to Perceive Targets via Historical Trajectories in Satellite Video Tracking
Jiahao Wang, Fang Liu, Licheng Jiao, Hao Wang, Shuo Li, Xinyi Wang, Lingling Li, Puhua Chen, Xu Liu
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 Adapter to SeqTrack backbone (ablation) | SeqTrack · SV248S test | SR: 44.3 -> 46.9 | Improved |
| add historical-trajectory text branch (ablation) | SeqTrack + Adapter · SV248S test | SR: 46.9 -> 48.8 | Improved |
| add Dynamic Scaling Alignment (DSA) (ablation) | HTTrack · SV248S test | SR: 48.8 -> 49.5 | Improved |
| add Cross-Modal Alignment (CMA) (ablation) | HTTrack · SV248S test | PR: 77.8 -> 79.0 | Improved |
| add Cross-Spatial Feature Mapper (CSFM) (ablation) | HTTrack · SV248S test | SR: 49.9 -> 50.9 | Improved |
| add current-frame label text constraint (Similarity) (ablation) | HTTrack · SV248S test | SR: 50.9 -> 51.5 | Improved |
| add CSFM (ablation) | HTTrack · SV248S test | PR: 79.0 -> 79.2 | No clear effect |
| use FLAN-T5 instead of Qwen2.5 as the language model (ablation) | HTTrack · SV248S test | SR: 50.2 -> 51.5 | Improved |
| use 5 instead of 10 historical trajectories (ablation) | HTTrack · SV248S test | SR: 48.9 -> 51.5 | Improved |
| use 5 instead of 1 historical trajectory (ablation) | HTTrack · SV248S test | SR: 50.6 -> 51.5 | Improved |
| add scaling factors in DSA (ablation) | HTTrack · SV248S test | SR: 50.6 -> 51.5 | Improved |
| use HTTrack instead of SVLPNet | satellite video tracking · SatSOT | success rate: 0.465 -> 0.529 | Improved |
Closest papers by meaning
- Foreground-Aware Token Routing Vision Transformer for Real-Time Satellite Video Tracking (ICML 2026 · similarity 0.86)
- HIPTrack: Visual Tracking with Historical Prompts (CVPR 2024 · similarity 0.84)
- SPMTrack: Spatio-Temporal Parameter-Efficient Fine-Tuning with Mixture of Experts for Scalable Visual Tracking (CVPR 2025 · similarity 0.83)
- Track-On: Transformer-based Online Point Tracking with Memory (ICLR 2025 · similarity 0.83)
- Autoregressive Queries for Adaptive Tracking with Spatio-Temporal Transformers (CVPR 2024 · similarity 0.83)
- Learning Adaptive and View-Invariant Vision Transformer for Real-Time UAV Tracking (ICML 2024 · similarity 0.83)
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.
