Jingkuan Song is a full professor with University of Electronic Science and Technology of China (UESTC). He joined Columbia University as a Postdoc Research Scientist
(2016-2017), and University of Trento as a Research Fellow (2014-2016). He obtained his PhD degree in 2014 from The University of Queensland (UQ), Australia (advised
by Prof. Heng Tao Shen). His research interest includes large-scale multimedia retrieval, image/video segmentation and image/video understading using hashing, graph learning and deep learning
techniques. He was the winner of the Best Paper Award in ICPR (2016, Mexico), Best Student Paper Award in Australian Database Conference (2017, Australia), and Best Paper Honorable Mention Award (2017, Japan). He is Guest Editor of TMM,
WWWJ and he is PC member of CVPR’18, MM'18, IJCAI'18, etc.
I am looking for highly motivated PhD students, Postdoctorals and Assistant Professors to conduct world-class research in my team. Please send your CV or enquiries to my email.
Jingkuan Song, Tao He, Lianli Gao, Xing Xu, Heng Tao Shen.
In AAAI, 2018, accepted.
Introduction: In this paper we propose a fast and effective Deep Region Hashing (DRH) approach for visual instance search in large scale datasets with an image patch as the query. Specifically, DRH is an end-to-end
deep neural network which consists of layers for feature extraction, object proposal, and hash code generation. DRH can generate nearly costfree region proposals because the region proposal network shares full-image convolutional feature
map. Also, the hash code generation layer learns binary codes to represent large scale images and regions to accelerate INS.
Introduction: It is worthwhile to highlight the main contributions of this proposed approach: 1) We introduce a novel hLSTMat framework which automatically decides when and where to use video visual information,
and when and how to adopt the language model to generate the next word for video captioning. 2) We propose a novel adjusted temporal attention mechanism which is based on temporal attention. Specifically, the temporal attention is used
to decide where to look at visual information, while the adjusted temporal model is designed to decide when to make use of visual information and when to rely on language model. A hierarchical LSTMs is designed to obtain low-level visual
information and high-level language context information. 3) Experiments on two benchmark datasets demonstrate that our method outperforms the state-of-the-art methods in both BLEU and METEOR.
Jingkuan Song, Lianli Gao, Mihai Marian Puscas, Feiping Nie, Fumin Shen, Nicu Sebe.
In MM, 2018, accepted.
Introduction: Our unsupervised video segmentation framework learns the similarity graph and cutting structure simultaneously to achieve the optimal segmentation results. We derive a novel and efficient algorithm to
solve this challenging problem. We utilized multiple cues of the superpixels and the weights of different cues are automatically learned. Furthermore, we calibrate the similarity of different superpixels based on their topology structures
to make them comparable. The proposed JGLVS achieves up to 11% improvement over the state-of-the-art baselines on the largest public dataset VSB100, which validates the effectiveness and efficiency of our approach.
Lei Zhu, Zi Huang, Xiaojun Chang, Jingkuan Song, Heng Tao Shen.Exploring Consistent Preferences: Discrete Hashing with Pair-Exemplar for Scalable Landmark Search.In Proceedings of 25th ACM International Conference on Multimedia (ACM MM), 2017, accepted.
Fumin Shen, Yadong Mu, Yang Yang, Wei Liu, Li Liu, Jingkuan Song, Heng Tao Shen.Classification by Retrieval: Binarizing Data and Classifier.SIGIR, 2017, accepted.[Best Paper Award Mention]
Xing Xu, Fumin Shen, Yang Yang, Dongxiang Zhang, Heng Tao Shen, Jingkuan Song.Matrix Tri-Factorization with Manifold Regularizations for Zero-shot Learning.CVPR, 2017, accepted.
Lianli Gao, Peng Wang, Jingkuan Song, Zi Huang, Jie Shao, Heng Tao Shen.Event Video Mashup: From Hundreds of Videos to Minutes of Skeleton.AAAI, 2017, accepted.
J. Yang, E. D. Angelini, P. P. Balte, E. A. Hoffman, J. H.M. Austin, B. Smith, Jingkuan. Song, R. G. Barr, A. F. Laine.Unsupervised Discovery of Spatially-Informed Lung Texture Patterns for Pulmonary Emphysema: The MESA COPD Study.In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2017, Accepted.
Jingkuan Song, Lianli Gao*, Xiaofeng Zhu, Nicu Sebe.Quantization based Hashing: A General Framework for Scalable Image and Video Retrieval.Pattern Recognition (PR), 2017, accepted.
Jingdong Wang, Ting Zhang, Jingkuan Song, Nicu Sebe, Heng Tao Shen.A Survey on Learning to Hash.IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2017, accepted.
Litao Yu, Zi Huang, Fumin Shen, Jingkuan Song, Heng Tao Shen, Xiaofang Zhou.Bilinear Optimized Product Quantization for Scalable Visual Content Analysis.IEEE Transactions on Image Processing (TIP), 2017, accepted.
Xiaofeng Zhu, Shichao Zhang, Yongyao Hu, Yonghua Zhu, Jingkuan Song.Local and Global Structure Preservation for Robust Unsupervised Spectral Feature Selection.IEEE Transactions on Knowledge and Data Engineering (TKDE), 2017, accepted.
Jingkuan Song, Lianli Gao, Mihai Marian Puscas, Feiping Nie, Fumin Shen, Nicu Sebe.Joint Graph Learning and Video Segmentation Via Multiple Cues and Topology Calibration.ACM Multimedia, 2016, accepted.
Dan Xu, Xavier Alameda-Pineda, Jingkuan Song, Elisa Ricci, and Nicu Sebe.Academic Coupled Dictionary Learning for Sketch-based Image Retrieval.ACM Multimedia, 2016, accepted.
Xing Xu, Fumin Shen, Zi Huang, Yang Yang, Li He, Jingkuan Song.Cross-modal Retrieval with Label Completion.ACM Multimedia(short), 2016, accepted.
Zhao Guo, Lianli Gao, Jingkuan Song, Xing Xu, Jie Shao, Heng Tao Shen.Attention-based LSTM with Semantic Consistency for Videos Captioning.ACM Multimedia(short), 2016, accepted.
Dan Xu, Jingkuan Song, Xavier Alameda-Pineda, Elisa Ricci, Nicu Sebe.Multi-Paced Dictionary Learning for Cross-Domain Retrieval and Recognition.ICPR, 2016, accepted.[Best Paper Award Mention]
Lianli Gao, Jingkuan Song, Feiping Nie, Fuhao Zou, Nicu Sebe and Heng Tao Shen.Graph-without-cut: An Ideal Graph Learning for Image Segmentation.AAAI, 2016, accepted.
Jingkuan Song, Lianli Gao, Feiping Nie, Yan Yan, Heng Tao Shen, and Nicu Sebe.Optimized Graph Learning using Partial Tags and Multiple Features for Image and Video Annotation.IEEE Transactions on Image Processing (TIP), 2016, accepted.
Jingkuan Song, Heng Tao Shen, Jianfeng Wang, Zi Huang, Nicu Sebe, Jingdong Wang.A Distance-Computation-Free Search Scheme for Binary Code Databases.Transactions on Multimedia (TMM), 2016, accepted.
Litao Yu, Yang Yang*, Zi Huang, Peng Wang, Jingkuan Song, Heng Tao Shen.Web Video Event Recognition by Semantic Analysis from Ubiquitous Documents.IEEE Transactions on Image Processing (TIP), 2016, accepted.
Fumin Shen, Xiang Zhou, Yang Yang, Jingkuan Song, Heng Tao Shen, Dacheng Tao.A Fast Optimization Method for General Binary Code Learning.IEEE Transactions on Image Processing (TIP), 2016, accepted.
Jingkuan Song, Lianli Gao, Yan Yan, Dongxiang Zhang and Nicu Sebe.Supervised Hashing with Pseudo Labels.ACM Multimedia(short), 2015, accepted.
Guoyu Lu, Yan Yan, Jingkuan Song, Nicu Sebe, Chandra Kambhamettu.Localize me Anywhere, Anytime: A Multi-task Point-Retrieval Approach.ICCV, 2015, accepted.
Lianli Gao, Jingkuan Song, Feiping Nie, Nicu Sebe, Heng Tao Shen.Optimal Graph Learning with Partial Tags and Multiple Features for Image and Video Annotation.CVPR, 2015, accepted.
Lianli Gao, Jingkuan Song, Yan Yan, Dongxiang Zhang and Nicu Sebe.Scalable Multimedia Retrieval by Deep Learning Hashing with Relative Similarity Learning.ACM Multimedia (short), 2015, accepted.
Fuhao Zou, Yunpeng Chen, Jingkuan Song, Ke Zhou, Yang Yang, Nicu Sebe.Compact Image Fingerprinting via Multiple Kernel Hashing.Transactions on Multimedia (TMM), 2015, accepted.
Jingkuan Song, Zi Huang, Heng Tao Shen and Yang Yang.Inter-Media Hashing for Large-Scale Multimedia Retrieval.ACM International Conference on Management of Data (SIGMOD), pages 785-796, 2013.
Jingkuan Song.Effective Hashing for Large-scale Multimedia Search.In ACM International Conference on Management of Data (SIGMOD), PhD Symposium, pages 55-60, 2013.
Jingkuan Song, Yi Yang, Zi Huang, Heng Tao Shen and Richang Hong.Multiple Feature Hashing for Real-time Large Scale Near-duplicate Video Retrieval.ACM Multimedia, 2011, accepted.
Jingkuan Song, Yi Yang, Xuelong Li, Zi Huang, Yang Yang.Robust Hashing with Local Models for Effective Similarity Search.IEEE Transactions on Cybernetics (TOC), 2014, accepted.
Jianfeng Wang, Jingdong Wang, Jingkuan Song, Xin-Shun Xu, Heng Tao Shen, Shipeng Li.Optimized Cartesian K-means.IEEE Transactions on Knowledge and Data Engineering (TKDE), 2014, accepted.
Yi Yang, Jingkuan Song, Zi Huang, Zhigang Ma, Nicu Sebe, Alexander G. Hauptmann.Multi-Feature Fusion via Hierarchical Regression for Multimedia Analysis.Transactions on Multimedia (TMM), 15(3):572-581, 2013.
Jingkuan Song, Yi Yang, Zi Huang, Heng Tao Shen, Jiebo Luo.Effective Multiple Feature Hashing for Large-scale Near-duplicate Video Retrieval.Transactions on Multimedia (TMM), 15(8):1-12, 2013.
2018 by Jingkuan Song. All Rights Reserved.