{"type":"ScholarlyArticle","title":"Tag-based Weakly-supervised Hashing for Image Retrieval","authors":[{"name":"Ziyu Guan","slug":"ziyu-guan","given":"Ziyu","family":"Guan","affiliation":"School of Information and Technology , Northwest University of China"},{"name":"Fei Xie","slug":"fei-xie","given":"Fei","family":"Xie","affiliation":"School of Information and Technology , Northwest University of China"},{"name":"Wanqing Zhao","slug":"wanqing-zhao","given":"Wanqing","family":"Zhao","affiliation":"School of Information and Technology , Northwest University of China"},{"name":"Xiaopeng Wang","slug":"xiaopeng-wang","given":"Xiaopeng","family":"Wang","affiliation":"School of Information and Technology , Northwest University of China"},{"name":"Long Chen","slug":"long-chen","given":"Long","family":"Chen","affiliation":"School of Information and Technology , Northwest University of China"},{"name":"Wei Zhao","slug":"wei-zhao","given":"Wei","family":"Zhao","affiliation":"School of Computer Science and Technology , Xidian University"},{"name":"Jinye Peng","slug":"jinye-peng","given":"Jinye","family":"Peng","affiliation":"School of Information and Technology , Northwest University of China"}],"abstract":"We are concerned with using user-tagged images to learn proper hashing functions for image retrieval. The benefits are two-fold: (1) we could obtain abundant training data for deep hashing models; (2) tagging data possesses richer semantic information which could help better characterize similarity relationships between images. However, tagging data suffers from noises, vagueness and incompleteness. Different from previous unsupervised or supervised hashing learning, we propose a novel weakly-supervised deep hashing framework which consists of two stages: weakly-supervised pre-training and supervised fine-tuning. The second stage is as usual. In the first stage, rather than performing supervision on tags, the framework introduces a semantic embedding vector (sem-vector) for each image and performs learning of hashing and sem-vectors jointly. By carefully designing the optimization problem, it can well leverage tagging information and image content for hashing learning. The framework is general and does not depend on specific deep hashing methods. Empirical results on real world datasets show that when it is integrated with state-of-art deep hashing methods, the performance increases by 8-10%.","doi":"10.24963/ijcai.2018/525","url":"https://ijcai.org/proceedings/2018/525","pdf":"https://ijcai.org/proceedings/2018/0525.pdf","markdown":"https://ijcai.org/proceedings/2018/525.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/525","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Multidisciplinary Topics and Applications","track":"Main track","number":525,"pages":{"first":3776,"last":3782},"keywords":["Machine Learning: Deep Learning","Multidisciplinary Topics and Applications: Information Retrieval"],"videos":[],"references":[{"position":1,"raw":"[Chua et al., 2009] Tatseng Chua, Jinhui Tang, Richang Hong, Haojie Li, Zhiping Luo, and Yantao Zheng. Nus- wide: a real-world web image database from national uni- versity of singapore. In CIVR, page 48, 2009.","title":"Nuswide: a real-world web image database from national university of singapore","authors":"Chua","year":"2009","venue":"CIVR","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Donoho and Tsaig, 2008] David L. Donoho and Yaakov Tsaig. Fast solution of ℓ 1 -norm minimization problems when the solution may be sparse. IEEE TIP, 54(11):4789- 4812, 2008.","title":"Fast solution of ℓ 1 -norm minimization problems when the solution may be sparse","authors":"David L Donoho, Yaakov Tsaig","year":"2008","venue":"IEEE TIP","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Gionis et al., 2000] Aristides Gionis, Piotr Indyk, Rajeev Motwani, et al. Similarity search in high dimensions via hashing. In VLDB, pages 518-529, 2000.","title":"Similarity search in high dimensions via hashing","authors":"Gionis","year":"2000","venue":"VLDB","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Huiskes et al., 2010] Mark J. Huiskes, Bart Thomee, and Michael S. Lew. New trends and ideas in visual concept detection: The mir flickr retrieval evaluation initiative. In ACM MIR, pages 527-536, 2010.","title":"New trends and ideas in visual concept detection: The mir flickr retrieval evaluation initiative","authors":"J Mark, Bart Huiskes, Michael S Thomee, Lew","year":"2010","venue":"ACM MIR","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Irie et al., 2014] Go Irie, Zhenguo Li, Xiaoming Wu, and Shihfu Chang. Locally linear hashing for extracting non- linear manifolds. In CVPR, pages 2115-2122, 2014.","title":"Locally linear hashing for extracting nonlinear manifolds","authors":"Irie","year":"2014","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Kulis and Darrell, 2009] Brian Kulis and Trevor Darrell. Learning to hash with binary reconstructive embeddings. In NIPS, pages 1042-1050, 2009.","title":"Learning to hash with binary reconstructive embeddings","authors":"Brian Kulis, Trevor Darrell","year":"2009","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Li and Tang, 2017] Zechao Li and Jinhui Tang. Weakly su- pervised deep matrix factorization for social image under- standing. IEEE TIP, 26(1):276-288, 2017.","title":"Weakly supervised deep matrix factorization for social image understanding","authors":"Li, Tang, Zechao Li, Jinhui Tang","year":"2017","venue":"IEEE TIP","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Lin et al., 2016] Kevin Lin, Jiwen Lu, Chu Song Chen, and Jie Zhou. Learning compact binary descriptors with un- supervised deep neural networks. In CVPR, pages 1183- 1192, 2016.","title":"Learning compact binary descriptors with unsupervised deep neural networks","authors":"Kevin Lin, Jiwen Lu, Chu Song Chen, Jie Zhou","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Liu et al., 2009] Dong Liu, Xiansheng Hua, Linjun Yang, Meng Wang, and Hongjiang Zhang. Tag ranking. In WWW, pages 351-360, 2009.","title":"Tag ranking","authors":"Dong Liu, Xiansheng Hua, Linjun Yang, Meng Wang, Hongjiang Zhang","year":"2009","venue":"WWW","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Liu et al., 2012] Wei Liu, Jun Wang, Rongrong Ji, Yugang Jiang, and Shihfu Chang. Supervised hashing with kernels. In CVPR, pages 2074-2081, 2012.","title":"Supervised hashing with kernels","authors":"Wei Liu, Jun Wang, Rongrong Ji, Yugang Jiang, Shihfu Chang","year":"2012","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Liu et al., 2016] Haomiao Liu, Ruiping Wang, Shiguang Shan, and Xilin Chen. Deep supervised hashing for fast image retrieval. In CVPR, pages 2064-2072, 2016.","title":"Deep supervised hashing for fast image retrieval","authors":"Liu","year":"2016","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Mikolov et al., 2013] Tomas Mikolov, Greg Corrado, Kai Chen, and Jeffrey Dean. Efficient estimation of word rep- resentations in vector space. In ICLR, pages 1-12, 2013.","title":"Efficient estimation of word representations in vector space","authors":"Mikolov","year":"2013","venue":"ICLR","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Niu et al., 2015] Yulei Niu, Zhiwu Lu, Songfang Huang, Peng Han, and Jirong Wen. Weakly supervised matrix factorization for noisily tagged image parsing. In IJCAI, pages 3749-3755, 2015.","title":"Weakly supervised matrix factorization for noisily tagged image parsing","authors":"Niu","year":"2015","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2015/527"},{"position":14,"raw":"[Qian et al., 2014] Xueming Qian, Xiansheng Hua, Yuanyan Tang, and Tao Mei. Social image tagging with diverse semantics. IEEE cybernetics, 44(12):2493-2508, 2014.","title":"Social image tagging with diverse semantics","authors":"Xueming Qian, Xiansheng Hua, Yuanyan Tang, Tao Mei","year":"2014","venue":"IEEE cybernetics","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Simonyan and Zisserman, 2015] Karen Simonyan and An- drew Zisserman. Very deep convolutional networks for large-scale image recognition. In NIPS, 2015.","title":"Very deep convolutional networks for large-scale image recognition","authors":"Karen Simonyan, Andrew Zisserman","year":"2015","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Sun et al., 2011] Aixin Sun, SouravS. Bhowmick, and Ju- nan Chong. Social image tag recommendation by concept matching. In ACM MM, pages 1181-1184, 2011.","title":"Social image tag recommendation by concept matching","authors":"Aixin Sun, Souravs Bhowmick, Junan Chong","year":"2011","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Tang and Li, 2017] Jinhui Tang and Zechao Li. Weakly- supervised multimodal hashing for scalable social image retrieval. IEEE TCSVT, PP(99):1-1, 2017.","title":"Weaklysupervised multimodal hashing for scalable social image retrieval","authors":"Li ; ] Jinhui Tang, Zechao Tang, Li","year":"2017","venue":"IEEE TCSVT","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Tang et al., 2009] Jinhui Tang, Shuicheng Yan, Richang Hong, Guo Jun Qi, and Tat Seng Chua. Inferring semantic concepts from community-contributed images and noisy tags. In ACM MM, pages 223-232, 2009.","title":"Inferring semantic concepts from community-contributed images and noisy tags","authors":"Jinhui Tang, Shuicheng Yan, Richang Hong, Guo Jun Qi, Tat Seng, Chua","year":"2009","venue":"ACM MM","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Weiss et al., 2008] Yair Weiss, Antonio Torralba, and Rob Fergus. Spectral hashing. In NIPS, pages 1753-1760, 2008.","title":"Spectral hashing","authors":"Yair Weiss, Antonio Torralba, Rob Fergus","year":"2008","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Wu et al., 2013] Lei Wu, Rong Jin, and Anil K. Jain. Tag completion for image retrieval. IEEE TPAMI, 35(3):716- 27, 2013.","title":"Tag completion for image retrieval","authors":"Lei Wu, Rong Jin, Anil K Jain","year":"2013","venue":"IEEE TPAMI","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Xia et al., 2015] Zhaoqiang Xia, Xiaoyi Feng, Jinye Peng, Jun Wu, and Jianping Fan. A regularized optimization framework for tag completion and image retrieval. Neu- rocomputing, 147(1):500-508, 2015.","title":"A regularized optimization framework for tag completion and image retrieval","authors":"Zhaoqiang Xia, Xiaoyi Feng, Jinye Peng, Jun Wu, Jianping Fan","year":"2015","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Yan et al., 2017] Xinyu Yan, Lijun Zhang, and Wujun Li. Semi-supervised deep hashing with a bipartite graph. In IJCAI, pages 3238-3244, 2017.","title":"Semi-supervised deep hashing with a bipartite graph","authors":"Xinyu Yan, Lijun Zhang, Wujun Li","year":"2017","venue":"IJCAI","doi":null,"url":null,"internal_url":"/proceedings/2017/452"},{"position":23,"raw":"[Zhang et al., 2015] Ruimao Zhang, Liang Lin, Rui Zhang, Wangmeng Zuo, and Lei Zhang. Bit-scalable deep hashing with regularized similarity learning for image retrieval and person re-identification. IEEE TIP, 24(12):4766-4779, 2015.","title":"Bit-scalable deep hashing with regularized similarity learning for image retrieval and person re-identification","authors":"Ruimao Zhang, Liang Lin, Rui Zhang, Wangmeng Zuo, Lei Zhang","year":"2015","venue":"IEEE TIP","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zhao et al., 2015] Fang Zhao, Yongzhen Huang, Liang Wang, and Tieniu Tan. Deep semantic ranking based hash- ing for multi-label image retrieval. In CVPR, pages 1556- 1564, 2015.","title":"Deep semantic ranking based hashing for multi-label image retrieval","authors":"Fang Zhao, Yongzhen Huang, Liang Wang, Tieniu Tan","year":"2015","venue":"CVPR","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Zhuang and Hoi, 2011] Jinfeng Zhuang and Steven C.H. Hoi. A two-view learning approach for image tag rank- ing. In WSDM, pages 625-634, 2011.","title":"A two-view learning approach for image tag ranking","authors":"Hoi ; ] Jinfeng Zhuang, Zhuang, C H Steven, Hoi","year":"2011","venue":"WSDM","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}