Abstract
In this paper, we propose a general approach for automatic segmentation, color-based retrieval and classification of garments in fashion store databases, exploiting shape and color information. The garment segmentation is automatically initialized by learning geometric constraints and shape cues, then it is performed by modeling both skin and accessory colors with Gaussian Mixture Models. For color similarity retrieval and classification, to adapt the color description to the users’ perception and the company marketing directives, a color histogram with an optimized binning strategy, learned on the given color classes, is introduced and combined with HOG features for garment classification. Experiments validating the proposed strategy, and a free-to-use dataset publicly available for scientific purposes, are finally detailed.
Highlights
We address the problem of automatic segmentation, color retrieval and classification of fashion garments
We propose a complete system for garment segmentation and color classification from images taken from on-line fashion stores
In order to quantify the effectiveness of the garment segmentation algorithm, we do not have a sufficient amount of manually drawn segmentations to be used as ground truth
Summary
We address the problem of automatic segmentation, color retrieval and classification of fashion garments. Upon the results of Garment Segmentation, color and type classification procedures are outlined in Sec. 3.4 and 3.6 respectively. A novel color histogram optimized on the color distribution of the dataset classes is employed for similarity retrieval. We demonstrate its effectiveness for automatic color based retrieval and garment classification. – our method employs a GMM color modeling to describe non interesting parts, such as skin and additional garments (not the item which is advertised by the image), and creates a segmentation by removing them;. – we propose a novel color descriptor which provides a discriminative summary of the color distribution of the region of interest; we provide a solution based on an extension of integral images to allow its fast computation;. – we provide a large dataset, used in our experiments, in order to allow the scientific community to test their solutions in comparison with our choices
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
