Sim2Real Bilevel Adaptation for Object Surface Classification using Vision-Based Tactile Sensors, dataset

This is the dataset used in the 'Sim2Real Bilevel Adaptation for Object Surface Classification using Vision-Based Tactile Sensors.' It contains 2 folders: train: a set of approximately 50,000 images converted via a diffusion model pre-trained on a small set of real images. test: a set of 1,152 images acquired in a real-world setting. Labels are assigned manually. Both folders contain a labels.csv file, showing the corresponding label for every image.