Research papers · Posted 2026.01.30 · Preprint
A Comparative Evaluation of Large Vision-Language Models for 2D Object Detection under SOTIF Conditions
A Comparative Evaluation of Large Vision-Language Models for 2D Object Detection under SOTIF Conditions
Overview
The study compares large vision-language models with specialized detectors on PeSOTIF traffic degradations. It analyzes recall and geometric precision trade-offs and discusses semantic models as potential high-level perception safety validators.
Author-supplied source information
Author-supplied information on arXiv.
8 pages, 11 figures
Version: v2