Research map / Detection and segmentation / Tracking
STMTrack: Template-Free Visual Tracking With Space-Time Memory Networks
Check 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 foreground-background label maps to the memory-branch input (ablation) | STMTrack with separate backbones · GOT-10k | AO: 0.568 -> 0.642 | Helped |
| add foreground-background label maps to the memory-branch input (ablation) | STMTrack with shared backbone · GOT-10k | AO: 0.591 -> 0.620 | Helped |
| use separate memory and query backbones instead of a shared backbone (ablation) | STMTrack with label maps · GOT-10k | AO: 0.620 -> 0.642 | Helped |
| use separate memory and query backbones instead of a shared backbone (ablation) | STMTrack with label maps · OTB-2015 | Success (AUC): 0.702 -> 0.719 | Helped |
| use separate memory and query backbones instead of a shared backbone (ablation) | STMTrack with label maps · TrackingNet | Success (AUC): 79.7 -> 80.3 | Helped |
| use separate memory and query backbones instead of a shared backbone (ablation) | STMTrack with label maps · LaSOT | Success (AUC): 0.593 -> 0.606 | Helped |
| use 3 rather than 1, 2, or 4 reference frames during training (ablation) | STMTrack · GOT-10k | AO: 0.629, 0.624, 0.642, 0.627 | Mixed |
| increase the inference memory size from 1 to 6 reference frames (ablation) | STMTrack · TrackingNet | Success (AUC): 79.1 -> 80.3 | Helped |
| increase inference memory size beyond 6 frames (ablation) | STMTrack · TrackingNet | Success (AUC): 80.3 -> 79.8 | Hurt |
Closest papers by meaning
- HIPTrack: Visual Tracking with Historical Prompts (CVPR 2024 · similarity 0.89)
- MIMTrack: In-Context Tracking via Masked Image Modeling (AAAI 2025 · similarity 0.88)
- SPMTrack: Spatio-Temporal Parameter-Efficient Fine-Tuning with Mixture of Experts for Scalable Visual Tracking (CVPR 2025 · similarity 0.88)
- Two-stream Beats One-stream: Asymmetric Siamese Network for Efficient Visual Tracking (AAAI 2025 · similarity 0.88)
- ODTrack: Online Dense Temporal Token Learning for Visual Tracking (AAAI 2024 · similarity 0.87)
- A Distractor-Aware Memory for Visual Object Tracking with SAM2 (CVPR 2025 · similarity 0.87)
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.