Context‐aware encoding for clothing parsing. Issue 12 (1st June 2019)

• Record Type: Journal Article
• Authors: Yoo, C.‐H.; Shin, Y.‐G.; Kim, S.‐W.; Ko, S.‐J.
• Is Part Of: Electronics letters. Volume 55:Issue 12(2019)
• Page Start: 692
• Page End: 693
• Publication Date: 2019-06-01
• Subjects: learning (artificial intelligence) -- clothing -- image classification -- neural nets -- image segmentation; clothing parsing -- clothing labels -- general scene semantic segmentation -- stylish match -- context‐aware outfit encoder -- COE -- clothes -- context‐aware encoding; Electronics -- Periodicals; 621.381
• DOI: 10.1049/el.2019.1213
• Languages: English
• ISSNs: 0013-5194

Abstract

Abstract : Clothing parsing is a special type of semantic segmentation in which each pixel is assigned with clothing labels. Unlike general scene semantic segmentation, stylish match (e.g. skirts + blouse, jeans + T‐shirt) is an important cue for recognising fine‐grained categories in clothing parsing. In this Letter, the authors propose a context‐aware outfit encoder (COE), as a side branch, that drives the convolutional neural network to take the stylish match into account for clothing parsing. The proposed COE provides information on matching clothes that can be utilised to improve the prediction accuracy of the base network significantly. Experimental results show that fully convolutional network and MobileNet with the COE improve the mean intersection of the union of those without the COE by 2.5 and 2.8%, respectively, on CFPD dataset.