Clothing Recognition Based on Clothing Co-occurrence Information and Multi-task Learning
HAO Zhifeng, LING Suiyi, WEN Wen, CAI Ruichu, YUAN Chang
Department of Computer Science, Guangdong University of Technology, Guangzhou 510006 Data Mining and Information Retrieval Laboratory, Guangdong University of Technology, Guangzhou 510006
Abstract Multi-task learning (MTL) ignores the influence of prior probability on the process of learning. Aiming at this problem, an approach for clothing recognition based on MTL and the co-occurrence information of the clothes categories (CA-MTL) is proposed. The prior constraint term is incorporated into the MTL model to integrate the co-occurrence information of the clothes categories. The original extended gradient method is modified correspondingly and the performance of the clothing classifiers is thus enhanced. Experimental results show that the average performance of CA-MTL outperforms those of single task learning, neural network and traditional multi-task learning. Furthermore, the training results of the proposed model are convenient for visualization and can be used for feature selection.
[1] KIAPOUR M H, YAMAGUCHI K, BERG A C, et al. Hipster Wars: Discovering Elements of Fashion Styles // Proc of the 25th IEEE European Conference on Computer Vision. Zurich, Switzerland: Springer International Publishing, 2014: 472-488. [2] YAMAGUCHI K, KIAPOUR M H, ORTIZ L E, et al. Parsing Clothing in Fashion Photographs // Proc of the 25th IEEE Confe-rence on Computer Vision and Pattern Recognition. Providence, USA: IEEE, 2012: 3570-3577. [3] LIU S, FENG J S, DOMOKOS C, et al. Fashion Parsing with Weak Color-Category Labels. IEEE Trans on Multimedia, 2014, 16(1): 253-265. [4] YAMAGUCHI K, KIAPOUR M H, BERG T L. Paper Doll Parsing: Retrieving Similar Styles to Parse Clothing Items // Proc of the International Conference on Computer Vision. Sydney, Australia: IEEE, 2013: 3519-3526. [5] LIU S, FENG J S, SONG Z, et al. Hi, Magic Closet, Tell Me What to Wear! // Proc of the 20th ACM International Conference on Multimedia. New York, USA: ACM, 2012: 619-628. [6] LIU S, SONG Z, LIU G C, et al. Street-to-Shop: Cross-Scenario Clothing Retrieval via Parts Alignment and Auxiliary Set // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Providence, USA: IEEE, 2012: 3330-3337. [7] HARA K, JAGADEESH V, PIRAMUTHU R. Fashion Apparel Detection: The Role of Deep Convolutional Neural Network and Pose-Dependent Priors[C/OL].[2015-09-20]. http://www.kotahara.com/uploads/1/8/2/0/18208959/egpaper_final.pdf. [8] LAO B, JAGADEESH K. Convolutional Neural Networks for Fa-shion Classification and Object Detection [EB/OL].[2015-09-07]. http://cs231n.stanford.edu/reports/BLAO_KJAG_CS231N_FinalPaperFashionClassification.pdf. [9] CHEN H Z, GALLAGHER A C, GIROD B. Describing Clothing by Semantic Attributes // Proc of the 12th European Conference on Computer Vision. Berlin, Germany: Springer-Verlag, 2012, III: 609-623. [10] YANG Y, RAMANAN D. Articulated Pose Estimation with Flexible Mixtures-of-Parts // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Providence, USA: IEEE, 2011: 1385-1392. [11] GIRSHICK R, MALIK J. Training Deformable Part Models with Decorrelated Features // Proc of the IEEE International Conference on Computer Vision. Sydney, Australia: IEEE, 2013: 3016-3023. [12] XUE Y, LIAO X J, CARIN L, et al. Multi-task Learning for Cla-ssification with Dirichlet Process Priors. Journal of Machine Lear-ning Research, 2007, 8: 35-63. [13] ZHOU J Y, CHEN J H, YE J P. MALSAR: Multi-task Learning via Structural Regularization[EB/OL].[2015-09-18]. http://www.public.asu.edu/~jye02/Software/MALSAR/Manual.pdf. [14] LIU J, JI S W, YE J P. Multi-task Feature Learning via Efficient L2, 1-Norm Minimization // Proc of the 25th Conference on Uncertainty in Artificial Intelligence. Arlington, USA: AUAI Press, 2009: 339-348. [15] JI S W, YE J P. An Accelerated Gradient Method for Trace Norm Minimization // Proc of the 26th Annual International Conference on Machine Learning. New York, USA: ACM, 2009: 457-464. [16] VAPNIK V N. The Nature of Statistical Learning Theory. New York, USA: Springer-Verlag, 2000. [17] ROTHER C, KOLMOGOROV V, BLAKE A. "Grabcut"-Interactive Foreground Extraction Using Iterated Graph Cuts. ACM Trans on Graphics (TOG), 2004, 23(3): 309-314. [18] CHANG C C, LIN C J. LIBSVM: A Library for Support Vector Machines. ACM Trans on Intelligent Systems and Technology (TIST), 2011, 2(3). DOI: 10.1145/1961189.1961199. [19] NGUYEN G H, BOUZERDOUM A, PHUNG S L. Learning Pa-ttern Classification Tasks with Imbalanced Data Sets // Yin P, eds. Pattern Recognition. 2009: 193-208. [20] MAZZA F, DA SILVA M P, LE CALLET P, et al. What Do You Think of My Picture? Investigating Factors of Influence in Profile Images Context Perception. Proc of theSPIE, 2015. DOI: 10.1117/12.2082817.
2016, Vol. 29 
