My current research interests are Image/Video Analysis and Human-Computer Interaction. The research group I led is now carrying out projects on Image/Video Matting, Image/Video Annotation, Image Retrieval via Sketch, TV Search, TV Program Classification and Detection, and User Behavior Analysis. We also explore Android/iOS Systems and Applications to make better human-computer interaction. My recent publications are listed in the end of this page.
Matting is one of the key techniques in many applications, which aims at accurately estimate foreground objects in image or video. Our research focuses on how we can do it effectively and efficiently. This work was done by Yiyang Gu and Huating Zhao.
We developed our HDR algorithm to calculate the amount of light in a given scene and uses that information to preserve details within the image, even in scenes with large variations in brightness, for more realistic looking images. This work was done by Shiqian Tao and Fan Zhang.
Blurred images can be found in many applications. How to deblur an image is a long studied issue. We focus on document image deblur algorithms. This work was done by Yanfei Wang and Wenjie Li.
Image Super-Resolution aims to enhance the resolution of an imaging system. This work was done by Jing Wang and Hao Ren.
Images can be searched in many ways. Using sketch as search input enables users to search freely as they draw. We also provide an online version of the program, which can be accessed here. This work was done by Zheming Wang and Jianwei Yuan.
Automatic Image/Video Annotation helps computer to understand what we, as human beings, see the world. Robust annotation algorithms can be used in many applications such as photo album organization, robots, image/video retrieval and etc.
TV has been the center of family entertainment for several decades. With the development of mobile devices and internet video sites, TV seems to be driven away from our sight. However, we believe with better interaction tools and search technologies, TV, as a unique big sharing screen for a whole family, is still not replaceable. We are collaborating with many leading companies in this industry to give TV a better future. This work was majorly done by batch of 2010 graduate students, led by Yiyang Gu and Yiqun Wu.
In order to gain better understanding of video sequences, it’s a challenging work to automatically classify video clips. Currently we are working on Advertisement Detection using classification algorithms. This work was majorly done by batch of 2010 graduate students, led by Huagui Zhang and Quanping Liu.
User behavior gives information on how people think of the product and what we can do to improve it. It’s vital to many user-centered applications. User behavior is not a static character. We are among the first batch of researchers who discovered the changing property of user behavior. This work was done by Ruijiang Li.
Android and iOS are definitely overwhelming in mobile platforms. We focus on deep understanding of Android OS as well as application development on both Android and iOS. This work was majorly done by batch of 2010 graduate students, led by Yiqun Wu and Xuejun Liu.
- Kuncheng Fang, Lian Zhou, Cheng Jin, Yuejie Zhang, Kangnian Weng, Tao Zhang, Weiguo Fan, Fully Convolutional Video Captioning with Coarse-to-Fine and Inherited Attention, AAAI 2019
- Yuanfang Guo, Xiaochun Cao, Rui Wang, Cheng Jin, A New Data Embedding Method with a New Data Embedding Domain for JPEG Images, 2018 IEEE Fourth International Conference on Multimedia Big Data (BigMM), 1-5, 2018
- Fei Huang, Cheng Jin, Yuejie Zhang, Kangnian Weng, Tao Zhang, Weiguo Fan, Sketch-based image retrieval with deep visual semantic descriptor, Pattern Recognition, Volume 76, April 2018, Pages 537-548
- Yong Cheng, Fei Huang, Lian Zhou, Cheng Jin, Yuejie Zhang and Tao Zhang, A Hierarchical Multimodal Attention-based Neural Network for Image Captioning, SIGIR 2017 (short paper)
- Fei Huang, Yong Cheng, Cheng Jin, Yuejie Zhang and Tao Zhang, Deep Multimodal Embedding Model for Fine-grained Sketch-based Image Retrieval, SIGIR 2017 (short paper)
- Fei Huang, Cheng Jin, Yuejie Zhang and Tao Zhang, Towards Sketch-based Image Retrieval with Deep Cross-modal Correlation, ICME 2017
- Gongze Cao, Yezhou Yang, Jie Lei, Cheng Jin, Yang Liu, Mingli Song, TripletGAN: Training Generative Model with Triplet Loss, arXiv preprint arXiv:1711.05084, 2017
- Cheng Jin, Chenjie Li, Zheming Wang, Yuejie Zhang, Tao Zhang, Sketch-based Image Retrieval with a Novel BoVW Representation, MMM 2016
- Fei Huang, Yong Cheng, Cheng Jin, Yuejie Zhang and Tao Zhang, Enhancing Sketch-Based Image Retrieval via Deep Discriminative Representation, ECAI 2016
- Yong Cheng, Fei Huang, Cheng Jin, Yuejie Zhang and Tao Zhang, A Novel Cross-Modal Topic Correlation Model for Cross-Media Retrieval, ECAI 2016
- Zhixin Liu, Cheng Jin, Yuejie Zhang, Tao Zhang, Automatic Naming of Speakers in Video via Name-Face Mapping, CCL 2016: 424-436
- Xingmeng Zhang, Cheng Jin, Yuejie Zhang, Tao Zhang, Image Tag Recommendation via Deep Cross-Modal Correlation Mining, CCL 2016: 437-449
- YongQing Liang, Cheng Jin, Yuejie Zhang, Salient Region Detection with Convex Hull Overlap, CoRR abs/1612.03284 (2016)
- Yong Cheng, Zhengxiang Cai, Rui Feng, Cheng Jin, Yuejie Zhang, Tao Zhang, Cross-Modal Image-Tag Relevance Learning for Social Images, ACM MultiMedia 2015 (short paper)
- Cheng Jin, Wenhui Mao, Ruiqi Zhang, Yuejie Zhang, Xiangyang Xue, Cross-Modal Image Clustering via Canonical Correlation Analysis, AAAI 2015
- Cheng Jin, Zheming Wang, Tianhao Zhang, Qinen Zhu, Yuejie Zhang, A Novel Visual-Region-Descriptor-based Approach to Sketch-based Image Retrieval, ICMR 2015
- Yong Cheng, Zhixin Liu, Yun Zhao, Cheng Jin, Yuejie Zhang, Tao Zhang, People News Search via Name-Face Association Analysis, ICMR 2015 (short paper)
- Shuai Ren, Cheng Jin, Chang Sun, Yuejie Zhang, Sketch-Based Image Retrieval via Adaptive Weighting. ICMR 2014
Yuejie Zhang, Wei Wu, Yang Li, Cheng Jin, Xiangyang Xue, Jianping Fan, Automatic Name-Face Alignment to Enable Cross-Media News Retrieval. IJCAI 2013
Ruijiang Li, Bin Li, Ke Zhang, Cheng Jin, Xiangyang Xue, Groupwise Constrained Reconstruction for Subspace Clustering, ICML 2012
Yao Lu, Wei Zhang, Cheng Jin, Xiangyang Xue, Learning Attention Map from Images, CVPR 2012
- Xisheng He, Zhe Wang, Cheng Jin, Yingbin Zheng, Xiangyang Xue, A simplified multi-class support vector machine with reduced dual optimization, Pattern Recognition Letters 2012
Cheng Jin, Chunlei Yang, Integrating Hierarchical Feature Selection and Classifier Training for Multi-Label Image Annotation, ACM SIGIR 2011
Ruijiang Li, Bin Li, Cheng Jin, Xiangyang Xue, Xingquan Zhu, Tracking User- Preference Varying Speed in Collaborative Filtering, AAAI 2011
Yuejie Zhang, Lei Cen, Cheng Jin, Xiangyang Xue and Jianping Fan, Learning Inter-Related Statistical Query Translation Models for English-Chinese Bi-Directional CLIR, IJCAI 2011
Yuejie Zhang, Lei Cen, Wei Wu, Cheng Jin and Xiangyang Xue, Fusion of Multiple Features and Supervised Learning for Chinese OOV Term Detection and POS Guessing, IJCAI 2011
Yiyang Gu, Cheng Jin, Xiangyang Xue, Quick matting: A matting method based on pixel spread and propagation, ICIP 2010
Yuejie Zhang, Lei Cen, Cheng Jin, Xiangyang Xue and Ning Zhou, Bilingual Query Translation and Expansion for Supporting More Effective Cross-Language Image Retrieval, ACM Multimedia 2010
Renzhong Wei, Hong Lu, Yingbin Zheng, Lei Cen, Cheng Jin, Xiangyang Xue, Weiguo Wu, How Context Helps: A Discriminative Codeword Selection Method for Object Detection, ICIP 2010
Jian-Feng Chen, Hong Lu, Renzhong Wei, Cheng Jin, Xiangyang Xue, An effective method for video genre classification, CIVR 2010