Research map / Vision-language and multimodal / Multimodal LLMs
Divert More Attention to Vision-Language Object Tracking
Mingzhe Guo, Zhipeng Zhang, Liping Jing, Haibin Ling, Heng Fan
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 ModaMixer (ablation) | SiamCAR · LaSOT | SUC: 50.7 -> 57.6 | Improved |
| add ASearch alone (ablation) | SiamCAR · LaSOT | SUC: 50.7 -> 52.1 | Improved |
| add ASearch on top of ModaMixer (ablation) | SiamCAR + ModaMixer · LaSOT | SUC: 57.6 -> 65.2 | Improved |
| asymmetric instead of symmetric searched network (ablation) | VLT_SCAR · LaSOT | SUC: 60.0 -> 63.9 | Improved |
| search post-processing block instead of fixed Shuffle block (ablation) | VLT_SCAR · LaSOT | SUC: 59.1 -> 63.9 | Improved |
| add residual connection in ModaMixer (ablation) | VLT_SCAR · LaSOT | SUC: 61.1 -> 63.9 | Improved |
| use BERT instead of GPT-2 as language model (ablation) | VLT_SCAR · LaSOT | SUC: 59.3 -> 63.9 | Improved |
| increase language-annotated training data (ablation) | VLT_SCAR · LaSOT | SUC: 58.9 -> 61.8 -> 63.9 | Improved |
| provide language description at inference (ablation) | VLT_SCAR · LaSOT | SUC: 53.4 -> 63.9 | Improved |
| use attribute words instead of sentence description (ablation) | VLT_SCAR · LaSOT | SUC: 63.9 -> 68.1 | Improved |
| apply contrastive loss at first backbone stage vs fourth (template branch) (ablation) | VLT_SCAR · LaSOT | SUC: 63.0 -> 66.0 | Improved |
| apply contrastive loss to both template and search branches (ablation) | VLT_SCAR · LaSOT | SUC: 65.5 -> 66.6 | Improved |
Closest papers by meaning
- ChatTracker: Enhancing Visual Tracking Performance via Chatting with Multimodal Large Language Model (NeurIPS 2024 · similarity 0.92)
- Unifying Visual and Vision-Language Tracking via Contrastive Learning (AAAI 2024 · similarity 0.92)
- MemVLT: Vision-Language Tracking with Adaptive Memory-based Prompts (NeurIPS 2024 · similarity 0.91)
- Mono3DVLT: Monocular-Video-Based 3D Visual Language Tracking (CVPR 2025 · similarity 0.89)
- VastTrack: Vast Category Visual Object Tracking (NeurIPS 2024 · similarity 0.89)
- MET-Bench: Multimodal Entity Tracking for Evaluating the Limitations of Vision-Language and Reasoning Models (ICML 2026 · similarity 0.88)
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.
