Research map / Vision-language and multimodal / Human and embodied modeling
GLAD: Generative Language-Assisted Visual Tracking for Low-Semantic Templates
Xingyu Luo, Cai Yidong, Jie Liu, Jie Tang, Gangshan Wu, Limin Wang
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 attention-pooled multimodal features instead of feature modulation with text (ablation) | GLAD-B256 · LaSOT | AUC: 69.5 -> 68.7 (-0.8) | Improved |
| use attention-pooled multimodal features instead of feature modulation with text (ablation) | GLAD-B256 · TNL2K | AUC: 59.7 -> 59.1 (-0.6) | Improved |
| use attention-pooled multimodal features instead of concatenating text in the decoder (ablation) | GLAD-B256 · LaSOT | AUC: 69.5 -> 68.4 (-1.1) | Improved |
| use attention-pooled multimodal features instead of concatenating text in the decoder (ablation) | GLAD-B256 · TNL2K | P: 62.2 -> 61.0 (-1.2) | Improved |
| include the 5th U-Net Transformer module features (ablation) | GLAD-B256 · LaSOT | AUC: 69.5 -> 66.5 (-3.0) | Improved |
| include the 5th U-Net Transformer module features (ablation) | GLAD-B256 · LaSOT | P: 74.2 -> 68.3 (-5.9) | Improved |
| include the 6th U-Net Transformer module features (ablation) | GLAD-B256 · LaSOT | AUC: 69.5 -> 68.6 (-0.9) | Improved |
| include the 7th U-Net Transformer module features (ablation) | GLAD-B256 · LaSOT | P: 74.2 -> 70.4 (-3.8) | Improved |
| additionally use the 10th U-Net expansion-path module (ablation) | GLAD-B256 · LaSOT | AUC: 69.5 -> 69.9 (+0.4) | Mixed |
| additionally use the 10th U-Net expansion-path module (ablation) | GLAD-B256 · TNL2K | P: 62.2 -> 63.5 (+1.3) | Improved |
| use diffusion U-Net generative fusion instead of a standard transformer fusion (ablation) | GLAD-B256 · LaSOT | AUC: 69.5 -> 67.2 (-2.3) | Improved |
| use diffusion U-Net generative fusion instead of a standard transformer fusion (ablation) | GLAD-B256 · LaSOT | P: 74.2 -> 69.7 (-4.5) | 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.
- ChatTracker — ChatTracker: Enhancing Visual Tracking Performance via Chatting with Multimodal Large Language Model (NeurIPS 2024) · better on 1 of 1 shared result
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
- Unifying Visual and Vision-Language Tracking via Contrastive Learning (AAAI 2024 · similarity 0.88)
- MVLM: Template-Free Tracking via Vision-Language Margin Confidence and Memory-Gated Tracking (CVPR 2026 · similarity 0.87)
- UTPTrack: Towards Simple and Unified Token Pruning for Visual Tracking (CVPR 2026 · similarity 0.87)
- Revisiting the Role of Language Priors in Vision-Language Models (ICML 2024 · similarity 0.86)
- ChatTracker: Enhancing Visual Tracking Performance via Chatting with Multimodal Large Language Model (NeurIPS 2024 · similarity 0.86)
- Generative-Based Fusion Mechanism for Multi-Modal 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.
