Research map / Detection and segmentation / Tracking
FeatTracker: Short- and Long-Range Temporal Feature Consistency for Robust Underwater Object Tracking
Jiaqing Li, Bin Lin, Chaocan Xue, Wu Ai, Qingping Zheng
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 hybrid training strategy to ODTrack (ablation) | ODTrack · VMAT | AUC (%): 54.66 -> 55.54 | Improved |
| add diffusion-like feature enhancement (ablation) | ODTrack + HTS · VMAT | AUC (%): 55.54 -> 57.12 | Improved |
| add LRTFC alone (ablation) | ODTrack + HTS · VMAT | AUC (%): 55.54 -> 56.62 | Improved |
| add SRTFC on top of DFE (ablation) | ODTrack + HTS + DFE · VMAT | AUC (%): 57.12 -> 58.73 | Improved |
| add LRTFC on top of DFE (ablation) | ODTrack + HTS + DFE · VMAT | AUC (%): 57.12 -> 57.57 | Improved |
| add SRTFC and LRTFC together on DFE (full model) (ablation) | ODTrack baseline · VMAT | AUC (%): 54.66 -> 61.25 | Improved |
| use diffusion-like enhancement instead of GAN (ablation) | feature enhancement · VMAT | AUC (%): 56.04 -> 61.25 | Improved |
| use diffusion-like enhancement instead of Transformer (ablation) | feature enhancement · VMAT | AUC (%): 57.41 -> 61.25 | Improved |
| use 700 diffusion steps instead of 600 (ablation) | DFE · VMAT | AUC (%): 60.01 -> 61.25 | Improved |
| use 700 diffusion steps instead of 900 (ablation) | DFE · VMAT | AUC (%): 59.60 -> 61.25 | Improved |
| use FIFO memory update instead of EMA (ablation) | SRTFC memory · VMAT | AUC (%): 60.33 -> 61.25 | Improved |
| use FIFO memory update instead of mean aggregation (ablation) | SRTFC memory · VMAT | AUC (%): 59.24 -> 61.25 | Improved |
Closest papers by meaning
- WebUOT-1M: Advancing Deep Underwater Object Tracking with A Million-Scale Benchmark (NeurIPS 2024 · similarity 0.90)
- When Trackers Date Fish: A Benchmark and Framework for Underwater Multiple Fish Tracking (AAAI 2026 · similarity 0.87)
- Long-term Tracking in the Wild: a Benchmark (ECCV 2018 · similarity 0.86)
- FlowTrack: Revisiting Optical Flow for Long-Range Dense Tracking (CVPR 2024 · similarity 0.86)
- WaterDiffusion: Learning a Prior-involved Unrolling Diffusion for Joint Underwater Saliency Detection and Visual Restoration (AAAI 2025 · similarity 0.85)
- VastTrack: Vast Category Visual Object Tracking (NeurIPS 2024 · 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.
