ibarcelo/

Automingo_dataset

Automingo-VQA is a vision-language dataset for safety-critical autonomous driving scene understanding. It is designed for Visual Question Answering (VQA), event-level reasoning, and structured analysis of ADAS-relevant traffic scenarios. The dataset focuses on real-world driving situations aligned with Euro NCAP-related safety scenarios, including cut-ins, traffic light transitions, vulnerable road user interactions, braking conflicts, construction zones… See the full description on the dataset page: https://huggingface.co/datasets/ibarcelo/Automingo_dataset.