How much labeled data do you actually need to deploy a parking occupancy system at a never-before-seen lot? A supervision study spanning CLIP zero-shot → ResNet-18 few-shot → full supervision on 432k parking space crops, with dataset annotation error discovery. Trained on NVIDIA A100 via IU Big Red 200.
python computer-vision deep-learning hpc pytorch supervised-learning resnet clip indiana-university few-shot-learning dataset-quality parking-occupancy pklot
-
Updated
Apr 6, 2026 - Python