ALMA WP7 Folding

Dataset accompanying the deliverables "D7.5 - Design of the benchmark tested for robotized ironing and garment folding - Initial" and "D7.6 - Design of the benchmark tested for robotized ironing and garment folding - Final" as part of the "ALMA: Human Centric Algebraic Machine Learning" project, CORDIS - EU research results Grant agreement ID: 952091. Compressed file composed by: 10 010 png images representing a 2.5 D image of a cloth with a single fold, all synthetically generated, size: 100 by 100 pixels. labels.txt file providing ground truth labels of pick and place information for unfolding, as comma separated values (image name, pick_x, pick_y, place_x, place_y). Intended for supervised learning, 10 000 png images with ground truth labels for training and cross-validation, and 10 png images with ground truth labels for additional testing and/or visualization.