Research map / Vision-language and multimodal / Cross-modal learning
SMSTracker: Tri-path Score Mask Sigma Fusion for Multi-Modal Tracking
Sixian Chan, Zedong Li, Wenhao Li, Shijian Lu, Chunhua Shen, Xiaoqin Zhang
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 score mask fusion (SMF) to the baseline (ablation) | OSTrack · LasHeR | PR: 51.5 -> 61.5 | Improved |
| add SMF to the baseline (ablation) | OSTrack · LasHeR | SR: 41.5 -> 49.2 | Improved |
| add SMF to the baseline (ablation) | OSTrack · DepthTrack | F-score: 52.9 -> 57.4 | Improved |
| add SMF to the baseline (ablation) | OSTrack · VisEvent | MPR: 69.5 -> 71.5 | Improved |
| add sigma interaction (SGI) to SMF variant (ablation) | SMSTracker · DepthTrack | F-score: 57.4 -> 60.8 | Improved |
| add drop key fine-tuning (DKF) to SMF+SGI (ablation) | SMSTracker · LasHeR | PR: 65.0 -> 70.3 | Improved |
| add DKF to SMF+SGI (ablation) | SMSTracker · DepthTrack | F-score: 60.8 -> 63.6 | Improved |
| add DKF to SMF+SGI (ablation) | SMSTracker · VisEvent | MPR: 73.6 -> 76.3 | Improved |
| use all three streams (RGB, modal, mix) instead of RGB only (ablation) | SMSTracker · LasHeR | PR: 68.5 -> 70.3 | Improved |
| use all three streams instead of RGB only (ablation) | SMSTracker · DepthTrack | F-score: 59.4 -> 63.6 | Improved |
| use mix stream alone vs RGB and modal streams (ablation) | SMSTracker · DepthTrack | F-score: 59.4 -> 61.6 | Improved |
| raise drop key ratio from 0 to 0.1 (ablation) | SMSTracker · LasHeR | PR: 64.3 -> 70.3 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SDSTrack — SDSTrack: Self-Distillation Symmetric Adapter Learning for Multi-Modal Visual Object Tracking (CVPR 2024) · better on 6 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Cross-Modality Distillation for Multi-Modal Tracking (TPAMI 2025 · similarity 0.90)
- Collaborating Vision, Depth, and Thermal Signals for Multi-Modal Tracking: Dataset and Algorithm (NeurIPS 2025 · similarity 0.90)
- Single-Model and Any-Modality for Video Object Tracking (CVPR 2024 · similarity 0.88)
- Bi-directional Adapter for Multimodal Tracking (AAAI 2024 · similarity 0.88)
- Cross-modulated Attention Transformer for RGBT Tracking (AAAI 2025 · similarity 0.87)
- SDSTrack: Self-Distillation Symmetric Adapter Learning for Multi-Modal Visual Object Tracking (CVPR 2024 · 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.
