ytj254/

VLM-CASE_carla_dataset

10,560 CARLA front-camera frames (1280×720) for driving scene-context classification, collected for the paper VLM-CASE: vision-language model enabled context-adaptive safety envelopes for anticipatory safe autonomous driving. Each frame is labeled with four scene-context fields:

Field Labels

road_surface dry, wet, snow

weather clear, heavy_rain, dense_fog

time_of_day day, night

illumination_assistance none, partial, strong… See the full description on the dataset page: https://huggingface.co/datasets/ytj254/VLM-CASE_carla_dataset.