提升交通视频理解效率与精度，用轻量方法实现细粒度时空分析。
STER-VLM: Spatio-Temporal With Enhanced Reference Vision-Language Models
- 分步处理时空信息，分离空间与时间特征提取
- 通过优选帧和参考驱动，提升动态场景理解能力
- 适合需要高效部署的智能交通系统应用
视觉语言模型（VLM）在自动化交通分析中表现出强大潜力，但现有方法通常计算开销大，难以实现细粒度的时空理解。本文提出高效框架STER-VLM，通过（1）图像描述分解以分别处理空间与时间信息；（2）基于最优视角筛选的时间帧选择策略，保证充分的时间上下文；（3）参考驱动机制捕捉细微运动与动态环境；（4）优化的视觉/文本提示设计。在WTS和BDD数据集上的实验表明，该方法显著提升了语义丰富度与交通场景解析能力。在AI City Challenge 2025 Track 2中取得55.655的测试成绩，验证了其在真实场景下资源高效且精准的交通分析能力。
原文摘要 · Abstract (English)
Vision-language models (VLMs) have emerged as powerful tools for enabling automated traffic analysis; however, current approaches often demand substantial computational resources and struggle with fine-grained spatio-temporal understanding. This paper introduces STER-VLM, a computationally efficient framework that enhances VLM performance through (1) caption decomposition to tackle spatial and temporal information separately, (2) temporal frame selection with best-view filtering for sufficient temporal information, and (3) reference-driven understanding for capturing fine-grained motion and dynamic context and (4) curated visual/textual prompt techniques. Experimental results on the WTS \cite{kong2024wts} and BDD \cite{BDD} datasets demonstrate substantial gains in semantic richness and traffic scene interpretation. Our framework is validated through a decent test score of 55.655 in the AI City Challenge 2025 Track 2, showing its effectiveness in advancing resource-efficient and accurate traffic analysis for real-world applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。