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.
@article{arxiv.2508.13470,
title = {STER-VLM: Spatio-Temporal With Enhanced Reference Vision-Language Models},
author = {Tinh-Anh Nguyen-Nhu and Triet Dao Hoang Minh and Dat To-Thanh and Phuc Le-Gia and Tuan Vo-Lan and Tien-Huy Nguyen},
journal= {arXiv preprint arXiv:2508.13470},
year = {2025}
}