From Segments to Scenes: Temporal Understanding in Autonomous Driving via Vision-Language Model

TL;DR

This paper introduces the TAD benchmark to evaluate vision-language models' ability to understand temporal dynamics in autonomous driving footage, revealing current limitations and proposing training-free solutions to enhance motion understanding.

Contribution

The paper presents the TAD benchmark focused on autonomous driving, evaluates existing models, and proposes two novel training-free methods to improve temporal understanding in this domain.

Findings

Current models show substandard accuracy on TAD due to poor motion understanding.

Scene-CoT and TCogMap improve accuracy by up to 17.72%.

TAD serves as a new benchmark for future research in autonomous driving temporal understanding.

Abstract

Temporal understanding in autonomous driving (AD) remains a significant challenge, even for recent state-of-the-art (SoTA) Vision-Language Models (VLMs). Prior work has introduced datasets and benchmarks aimed at improving temporal reasoning, but these have emphasized other video content, including sports, cooking, and movies. No existing benchmark focuses exclusively on the unique challenges of temporal understanding in ego-centric AD footage. To fill this gap, the Temporal Understanding in Autonomous Driving (TAD) benchmark is presented, which evaluates VLMs' ability to capture the dynamic relationships between actions in AD. TAD comprises nearly 6,000 question-answer (QA) pairs, spanning 7 human-designed tasks. In addition, an evaluation is performed that consists of 9 closed- and open-source generalist models as well as SoTA AD specialist models. When applied to TAD, current SoTA models demonstrated substandard accuracies, largely due to imperfect fine-grained motion understanding. To improve motion understanding and overall accuracy on TAD, two novel training-free solutions are proposed: Scene-CoT, that leverages Chain-of-Thought (CoT) and TCogMap, which incorporates an ego-centric temporal cognitive map. The proposed approaches are integrated with existing VLMs and improve average accuracy on TAD by up to 17.72%. By introducing TAD, benchmarking multiple SoTA models, and proposing effective enhancements, this work aims to catalyze future research on temporal understanding in AD. The benchmark and evaluation code are available at \href{https://huggingface.co/datasets/vbdai/TAD}{Hugging Face} and \href{https://github.com/vbdi/tad_bench}{Github}, respectively.