Unimore logo AImageLab
Back to the research area

Garment Selection and Color Classification

Internet shopping has grown incredibly in the last years, and fashion created an interesting application field for image understanding and retrieval, since hundreds of thousands images of clothes constitute a challenging dataset to be used for automatic or semi-automatic segmentation strategies, color analysis, texture analysis, similarity retrieval, automatic piece of clothing classification and so on. We addressed the problem of automatic segmentation, color retrieval and classification of fashion garments.


Garment Classification

Depending on the availability of manually photo retouched images (as often happens) background removal is performed with simple thresholding or a more sophisticated approach. A Random Forest classification on projection features is used to classify the product category , while Gaussian Mixture Models (GMM) are exploited to select the interesting piece of clothing of the picture, with an automated initialization procedure. A novel color histogram specifically optimized on the color distribution of the dataset classes is finally employed for similarity retrieval.

Publications

1 Manfredi, Marco; Grana, Costantino; Calderara, Simone; Cucchiara, Rita "A complete system for garment segmentation and color classification" MACHINE VISION AND APPLICATIONS, vol. 25, pp. 955 -969 , 2014 | DOI: 10.1007/s00138-013-0580-3 Journal
2 Grana, Costantino; Calderara, Simone; Borghesani, Daniele; Cucchiara, Rita "Learning Non-Target Items for Interesting Clothes Segmentation in Fashion Images" Proceedings of the 21st International Conference on Pattern Recognition (ICPR 2012), Tsukuba Science City, Japan, pp. 3317 -3320 , Nov 11-15, 2012 Conference
3 Grana, Costantino; Borghesani, Daniele; Cucchiara, Rita "Class-based color bag of words for fashion retrieval" Proceedings of the 2012 IEEE International Conference on Multimedia and Expo, Melbourne, Australia, pp. 444 -449 , Jul 9-13, 2012 | DOI: 10.1109/ICME.2012.13 Conference