EPIPTrack: Rethinking Prompt Modeling with Explicit and Implicit Prompts for Multi-Object Tracking
This work improves multi-object tracking for applications in surveillance and autonomous systems by enhancing adaptability and reducing hallucinations, though it is incremental as it builds on existing multimodal tracking methods.
The paper tackles the problem of multi-object tracking by addressing the limitations of static textual descriptions from large language models, which lack adaptability to real-time changes and are prone to hallucinations, and proposes EPIPTrack, a framework using explicit and implicit prompts for dynamic target modeling, achieving superior performance on benchmarks like MOT17, MOT20, and DanceTrack.
Multimodal semantic cues, such as textual descriptions, have shown strong potential in enhancing target perception for tracking. However, existing methods rely on static textual descriptions from large language models, which lack adaptability to real-time target state changes and prone to hallucinations. To address these challenges, we propose a unified multimodal vision-language tracking framework, named EPIPTrack, which leverages explicit and implicit prompts for dynamic target modeling and semantic alignment. Specifically, explicit prompts transform spatial motion information into natural language descriptions to provide spatiotemporal guidance. Implicit prompts combine pseudo-words with learnable descriptors to construct individualized knowledge representations capturing appearance attributes. Both prompts undergo dynamic adjustment via the CLIP text encoder to respond to changes in target state. Furthermore, we design a Discriminative Feature Augmentor to enhance visual and cross-modal representations. Extensive experiments on MOT17, MOT20, and DanceTrack demonstrate that EPIPTrack outperforms existing trackers in diverse scenarios, exhibiting robust adaptability and superior performance.