How Well Do Vision-Language Models Understand Sequential Driving Scenes? A Sensitivity Study
This work addresses the need for systematic evaluation of VLMs in autonomous driving tasks, providing a framework for sensitivity analysis, but it is incremental as it builds on existing datasets and methods.
The study tackled the problem of evaluating Vision-Language Models (VLMs) on sequential driving scenes, revealing that even top models achieve only 57% accuracy, falling short of human performance at 65% and exposing significant gaps in understanding vehicle dynamics and temporal relations.
Vision-Language Models (VLMs) are increasingly proposed for autonomous driving tasks, yet their performance on sequential driving scenes remains poorly characterized, particularly regarding how input configurations affect their capabilities. We introduce VENUSS (VLM Evaluation oN Understanding Sequential Scenes), a framework for systematic sensitivity analysis of VLM performance on sequential driving scenes, establishing baselines for future research. Building upon existing datasets, VENUSS extracts temporal sequences from driving videos, and generates structured evaluations across custom categories. By comparing 25+ existing VLMs across 2,600+ scenarios, we reveal how even top models achieve only 57% accuracy, not matching human performance in similar constraints (65%) and exposing significant capability gaps. Our analysis shows that VLMs excel with static object detection but struggle with understanding the vehicle dynamics and temporal relations. VENUSS offers the first systematic sensitivity analysis of VLMs focused on how input image configurations - resolution, frame count, temporal intervals, spatial layouts, and presentation modes - affect performance on sequential driving scenes. Supplementary material available at https://V3NU55.github.io