Research map / Detection and segmentation / Tracking
Hierarchical Attentive Recurrent Tracking
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 |
|---|---|---|---|
| Use HART instead of the prior tracker | KTH pedestrian tracking · KTH test | average IoU: 55.03% -> 77.11% | Helped |
| Use HART instead of the prior tracker | KITTI tracking · KITTI, 60 time-steps | Avg. IoU: 0.14 -> 0.81 | Helped |
| Add appearance attention and its loss, along with the omitted attention-parameter loss (ablation) | Spatial Att · KITTI, 60 time-steps | Avg. IoU: 0.60 -> 0.81 | Helped |
| Add the appearance-attention loss (ablation) | App Att · KITTI, 60 time-steps | Avg. IoU: 0.78 -> 0.81 | Helped |
| Add auxiliary tasks, including spatial-attention and appearance-attention losses | HART · real-world object tracking | training convergence | Helped |
| Use tracking loss alone, without direct spatial-attention supervision (ablation) | HART · object tracking | training convergence | Hurt |
| Add appearance-attention loss to guide object-specific masks (ablation) | HART · object tracking under occlusion | tracking performance | Helped |
Closest papers by meaning
- Deep Attentive Tracking via Reciprocative Learning (NeurIPS 2018 · similarity 0.87)
- Tracking objects that change in appearance with phase synchrony (ICLR 2025 · similarity 0.85)
- Recurrently Target-Attending Tracking (CVPR 2016 · similarity 0.85)
- Exploiting Spatial Invariance for Scalable Unsupervised Object Tracking (AAAI 2020 · similarity 0.85)
- Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision (NeurIPS 2025 · similarity 0.83)
- Beyond Accuracy: Tracking more like Human via Visual Search (NeurIPS 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.