Explicit-Implicit Prompt Injection and Semantic-Guided Latent LoRA for Vision-Language Tracking
- 12 December 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 33 (10709908) , 376-380
Abstract
Prompt-based learning has shown promise in visual-language tracking (VLT), yet existing methods often rely on either explicit or implicit prompting alone, limiting fine-grained cross-modal alignment. Moreover, Low-Rank Adaptation (LoRA) -based fine-tuning in prior work typically focuses on visual-only adaptation, overlooking language semantics. To address these issues, we propose a unified VLT framework that integrates Explicit-Implicit Prompt Injection (EIPI) and Semantic-Guided Latent LoRA (SGLL). EIPI introduces semantic prompts to facilitate robust and context-sensitive target modeling through two pathways. The explicit prompts are constructed by interact between multi-modal target representations with the search region, while implicit prompts are learned from linguistic features via a lightweight bottleneck network. Then, SGLL extends standard LoRA by introducing learnable queries in the latent space, allowing residual modulation based on language-visual semantics without retraining the full model. This dual design yields a parameter-efficient tracker with strong cross-modal adaptability. Extensive experiments show our method outperforms prior prompt-based approaches while maintaining high efficiency.Keywords
Funding Information
- National Natural Science Foundation of China (62441231)
- Key R&D projects of Liaoning Province (2024JH2/102500015)
- Fundamental Research Funds for the Central Universities of Ministry of Education (N25BSS034)
- Key Research Development and Promotion Special Project of Henan Province (232102210094)
- Henan Province International Science and Technology Cooperation Project (242102520036)
This publication has 24 references indexed in Scilit:
- Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance2024
- Multi-modal visual tracking based on textual generationInformation Fusion, 2024
- Context-Aware Integration of Language and Visual References for Natural Language Tracking2024
- OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning2024
- Joint Visual Grounding and Tracking with Natural Language Specification2023
- Transformer vision-language tracking via proxy token guided cross-modal fusionPattern Recognition Letters, 2023
- Joint Feature Learning and Relation Modeling for Tracking: A One-Stream Framework2022
- Towards More Flexible and Accurate Object Tracking with Natural Language: Algorithms and Benchmark2021
- LaSOT: A High-Quality Benchmark for Large-Scale Single Object Tracking2019
- Generation and Comprehension of Unambiguous Object Descriptions2016