International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064


Downloads: 117 | Views: 273

Review Papers | Information Technology | India | Volume 4 Issue 7, July 2015 | Popularity: 6.8 / 10


     

Review: Parsing Clothes by Giving Weak Supervision using Pose Estimation in Fashion Photographs

Mayuri P. Waghmode, Umesh B. Chavan


Abstract: In this paper address an effective method for parsing clothing in fashion photographs. The problem is challenging due to clothing look, layering, fashion, and body shape variation and pose. Presenting the problem of automatically parsing the fashion photographs with weak supervision from the user-generated colour-category labels such as grey-shirt and Black-skirt. Due to immense diversity of fashion items this problem has become very demanding. To solve this problem, the combination of the super pixel-level category classifier learning module, human pose identification module, the Marcov Random Field based colour and category implication module to create multiple well performing category classifiers, which can be applied directly to parse the garments and other items in the images. All the training images are parsed with colour-category tags and the human poses of the images are estimated during the learning phase. Finally, resulting improved pose estimation, providing parsing results with classifier for further use during test.


Keywords: Markov Random Fields, Fashion Parsing, Weakly supervised learning, SVM


Edition: Volume 4 Issue 7, July 2015


Pages: 302 - 305



Make Sure to Disable the Pop-Up Blocker of Web Browser




Text copied to Clipboard!
Mayuri P. Waghmode, Umesh B. Chavan, "Review: Parsing Clothes by Giving Weak Supervision using Pose Estimation in Fashion Photographs", International Journal of Science and Research (IJSR), Volume 4 Issue 7, July 2015, pp. 302-305, https://www.ijsr.net/getabstract.php?paperid=SUB156289, DOI: https://www.doi.org/10.21275/SUB156289