Research map / Vision-language and multimodal / Cross-modal learning
Modality-missing RGBT Tracking: Invertible Prompt Learning and High-quality Benchmarks
Andong Lu, Chenglong Li, J. Zhao, Jin Tang, Bin Luo
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 |
|---|---|---|---|
| use IPer in all layers instead of the first layer only (ablation) | IPL · RGBT234-Miss | PR: 80.6 -> 82.0 | Improved |
| use IPer in all layers instead of the first layer only (ablation) | IPL · LasHeR245-Miss | PR: 59.7 -> 61.7 | Improved |
| use invertible prompter instead of LoRA (ablation) | IPL · RGBT234-Miss | PR: 79.1 -> 82.0 | Improved |
| use invertible prompter instead of LoRA (ablation) | IPL · LasHeR245-Miss | PR: 59.1 -> 61.7 | Improved |
| add task alignment loss (ablation) | IPL · RGBT234-Miss | PR: 80.3 -> 82.0 | Improved |
| add task alignment loss (ablation) | IPL · LasHeR245-Miss | PR: 60.2 -> 61.7 | Improved |
| add IPer and TA over Copy baseline (ablation) | Base w/ Copy · RGBT234-Miss | PR: 74.8 -> 82.0 | Improved |
| use concatenation fusion instead of summation (ablation) | baseline tracker · RGBT234-Miss | PR: 75.4 -> 77.2 | Improved |
| use Transformer fusion instead of concatenation (ablation) | baseline tracker · RGBT234-Miss | PR: 77.2 -> 74.8 | Hurt |
| use shared-specific extractor instead of shared extractor (ablation) | baseline tracker · RGBT234-Miss | PR: 74.8 -> 75.4 | Improved |
| add shared-specific architecture (no IPer) in complete modality (ablation) | Base · RGBT234 | PR: 86.4 -> 88.3 | Improved |
| train with missing-modality data (ablation) | TBSI · RGBT234-Miss | PR: 76.1 -> 77.7 | Improved |
Closest papers by meaning
- Collaborating Vision, Depth, and Thermal Signals for Multi-Modal Tracking: Dataset and Algorithm (NeurIPS 2025 · similarity 0.89)
- Temporal Adaptive RGBT Tracking with Modality Prompt (AAAI 2024 · similarity 0.88)
- Cross-modulated Attention Transformer for RGBT Tracking (AAAI 2025 · similarity 0.88)
- Single-Model and Any-Modality for Video Object Tracking (CVPR 2024 · similarity 0.88)
- RGBT Tracking via All-layer Multimodal Interactions with Progressive Fusion Mamba (AAAI 2025 · similarity 0.88)
- Bi-directional Adapter for Multimodal Tracking (AAAI 2024 · similarity 0.86)
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.
