Thispaperdescribesamethodofclothingclassificationusingasingleimage.Themethodisintendedtobeused for building autonomous systems, that can recognize casuallythrownordinaryclothing.AsetofGaborfiltersisapplied to an input image, and image features invariant to translation, rotation and scale are then generated. In this paper, we propose descriptions of the features, focusing on clothing fabrics, wrinkles, and cloth overlaps. In addition, to deal with situations involving clumped clothing, the description is extended by combining with superpixel representation. Experimentsusingastatedescriptionandclassificationusing real clothing demonstrate the effectiveness of the proposed method.

EB00000003554K

Available

• A method of classifying crumpled clothing based on image features derived from clothing fabrics and wrinkles