Research map / Vision-language and multimodal / Human and embodied modeling
Beyond Explicit Language: Plug-and-Play Visual-to-Linguistic Modeling Toward General Object Tracking
Kaiyang Lan, Ying Cui, Chenchen Jing, Jianwei Zheng, Dongyan Guo
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 |
|---|---|---|---|
| plug in TIMI (textual inversion + semantic injection) | SeqTrack-B256 · LaSOT | AUC: 69.9 -> 71.0 | Improved |
| plug in TIMI | MCITrack-B224 · LaSOT | AUC: 75.3 -> 76.1 | Improved |
| plug in TIMI | DUTrack-B256 · LaSOT | AUC: 73.0 -> 73.6 | Improved |
| plug in TIMI | DUTrack-B256 · TNL2K | AUC: +3.5 AUC, to 67.2 | Improved |
| plug in TIMI | MCITrack-B224 · TrackingNet | AUC / PNorm: same performance | No clear effect |
| plug in TIMI | SeqTrack-B256 · TrackingNet | AUC: 83.3 -> 83.4 | No clear effect |
| plug in TIMI | SeqTrack-B256 / MCITrack-B224 · GOT-10K | AO | Improved |
| remove natural-language description (ablation) | DUTrack-B256 · LaSOT | AUC: 73.0 -> 72.3 | Hurt |
| replace text with pseudo-description (ablation) | DUTrack-B256 · LaSOT | AUC: 73.0 -> 73.4 | Improved |
| add pseudo-description alongside text (ablation) | DUTrack-B256 · LaSOT | AUC: 73.0 -> 73.6 | Improved |
| use CLIP text encoder rather than none (ablation) | SeqTrack-B256* · LaSOT | AUC: 70.5 -> 71.0 | Improved |
| use CLIP text encoder rather than BERT (ablation) | SeqTrack-B256* · LaSOT | AUC: 70.4 -> 71.0 | Improved |
Closest papers by meaning
- ChatTracker: Enhancing Visual Tracking via LLM-Driven Iterative Description Refinement (TPAMI 2026 · similarity 0.92)
- Generalizing Multiple Object Tracking to Unseen Domains by Introducing Natural Language Representation (AAAI 2023 · similarity 0.91)
- ChatTracker: Enhancing Visual Tracking Performance via Chatting with Multimodal Large Language Model (NeurIPS 2024 · similarity 0.91)
- Unifying Visual and Vision-Language Tracking via Contrastive Learning (AAAI 2024 · similarity 0.91)
- Mono3DVLT: Monocular-Video-Based 3D Visual Language Tracking (CVPR 2025 · similarity 0.87)
- MVLM: Template-Free Tracking via Vision-Language Margin Confidence and Memory-Gated Tracking (CVPR 2026 · 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.
