Learning and Vision Lab, ECE, NUS
Our lab members, including Tao Wang, Yu Li, Bingyi Kang, Jun Hao Liew, Sheng Tang, and Jiashi Feng, won the 1st Place Entry award at the 1st ICCV 2019 COCO Large Vocabulary Instance Segmentation (LVIS) Challenge at 10/27/2019 at Seoul, Korea.
A classification calibration method for long-tail instance segmentation was proposed to improve the inaccurate proposal classification for tail classes. The authors calibrate the prediction of classification head with a specifically designed twin head which is trained with proposal-level class balance sampling strategy. Without much additional cost and modification, this method improves the performance of baseline by a large margin on tail classes. Our team is one of the co-winners of the challenge, and we surpass the runner-up team by 2.7% in mAP.
Four post-doctoral research fellow positions are available now (starting as early as Oct 2019) at the Learning and Vision Group. The postdocs will have the opportunity to work with world class researchers on cutting edge problems with profound academic and societal impact. In particular, they will work on deep learning, and/or their applications for computer vision problems such as object detection, segmentation and human pose estimation. Competitive compensation will be offered to highly competitive candidates.
The ideal candidate should meet the following requirements: (a) Possess or be in the process of completing a PhD degree in the area of machine learning, deep learning or computer vision; (b) Have a strong research record, demonstrated by having been published in prestigious journals and conferences (e.g., ICML/NIPS/CVPR/ECCV/ICCV/TPAMI); (c) Have strong programming skills; (d) Be self-motivated and good at team working.
Interested applicants please email your cv to email@example.com. [29/07/2019]
Our lab members won the "Best Student Paper Award" at the 26th ACM International Conference on Multimedia which was held during 22-26 Oct, 2018, at Seoul, Republic of Korea. The awarded work by Jian Zhao, Jianshu Li, Yu Cheng, Terence Sim, Shuicheng Yan, Jiashi Feng was titled "Understanding Humans in Crowded Scenes: Deep Nested Adversarial Learning and A New Benchmark for Multi-Human Parsing". Congratulations!
Attached are Zhao Jian and Jianshu giving presentation at the conference (image 1) and them receiving the award with Jiashi (image 2). See image in new tab to zoom in. [30/10/2018]
Dr. Feng Jiashi, Head of our lab, has been listed as one of the ten young outstanding innovators from the Asia Pacific region (Southeast Asia, Australia, New Zealand and Taiwan) on the MIT Technology Review's "Innovators Under 35" Asia 2018 list.
Dr Feng will automatically become a candidate and potential finalist for the global "Innovators Under 35" list celebrated in the fall of 2018 in Boston, USA. He will deliver a three-minute presentation at EmTach Asia in Singapore on 30-31 January 2018.
NUS-Qihoo 360 Joint Lab achieved NO. 1 on ILSVRC 2017 Object Localization Task, with all competition tasks within Top 3! This project is collaborated by NUS LV group (Yunpeng Chen, Jianan Li, Yunchao Wei, Huaxin Xiao, Jianshu Li, Mengdan Zhang, Xuecheng Nie, Xiaojie Jin, Jiashi Feng) and Qihoo 360 AI institute (Jian Dong, Shuicheng Yan). See the ILSVRC 2017 official website for official leaderboard. [20/07/2017]
The LV team achieved the top performance on the ICCV 2017 MS-Celeb-1M Large-Scale Face Recognition Hard Set / Random Set / Low-Shot Learning Challenge! This project is collaborated by NUS LV group (Feng Jiashi, Zhao Jian, Wang Zhecan) and Panasonic R&D Center Singapore (Shen Shengmei, Xu Yan, Cheng Yu, Xiong Lin, Jayashree Karlekar, Sugiri Pranata), outperforming the 2nd-best by a large margin. See the MS-Celeb-1M official website for official leaderboard, and also report of ECE, NUS. [19/07/2017]
The LV team achieved the top-2 performance on the CVPR 2017 Visual Understanding of Humans in Crowd Scene & the 1st Look into Person (LIP) Challenge Human Parsing and Pose Task! This project is supervised by Prof. Yan Shuicheng, collaborated by Zhao Jian, Nie Xuecheng, Xiao Huaxin, Chen Yunpeng, Li Jianshu, (the first three authors are with equal contributions), outperforming the 3rd-best by a large margin. See the CVPR 2017 official website for details. [05/06/2017]
The LV team achieved the top performance on the National Institute of Standards and Technology (NIST) 2017 IARPA Janus Benchmark A (IJB-A) Face Verification Challenge and Face Identification Challenge!
This project is collaborated by NUS LV group (Feng Jiashi, Zhao Jian) and Panasonic R&D Center Singapore (Shen Shengmei, Xiong Lin, Jayashree Karlekar, Sugiri Pranata), outperforming the 2nd-best by a large margin. See the NIST official website for official report. We also release a technical report on arxiv titled "A Good Practice Towards Top Performance of Face Recognition: Transferred Deep Feature Fusion". [26/04/2017]
The paper "Computational Face Reader" by Xiangbo Shu, Liyan Zhang, Jinhui Tang, Guo-Sen Xie, and Shuicheng Yan, won the Best Student Paper Award in the 22nd Anniversary International Conference on MultiMedia Modeling held in Miami, USA during Jan 4 -- Jan 6, 2016. [08/01/2016]
NUS-LV's latest Facial Landmark Detection and Tracking framework won the first prize on the Category-2 and Category-3 in the first facial landmark tracking competition, 300 Videos in the Wild (300-VW): Facial Landmark Tracking in-the-Wild Challenge & Workshop (ICCV 2015). See the 300-VW competition at "http://ibug.doc.ic.ac.uk/resources/300-VW/" and the manuscript "Facial landmark detection via progressive initialization" by Shengtao Xiao, Shuicheng Yan and Ashraf Kassim. [21/12/2015]
The Orgnization Committee of the Workshop on Transferring and Adapting Source Knowledage (TASK) in Computer Vision (CV), held in Santiago de Chile on December 11th, 2015, awards the Best Paper Prize to the manuscript titled "Return of Frustratingly Easy Domain Adaptation" by Baochen Sun, Jiashi Feng and Kate Saenko. [21/12/2015]
Assistant Prof. FENG Jiashi would like to recruit two postdoctoral researchers and two research assistants. Please find more details in Recuitment. [16/11/2015]
Group member, Mr. LIN Min, received theBest Student Paper Award (Golden Prize) from Pattern Recognition and Machine Intelligence Association, Singapore (PREMIA) for the paper "Network in Network" in ICLR14. [20/03/2015]
NUS-LV's latest face recognition solution based on deeper network-in-network reaches 99.70% on the popular LFW face dataset (joint efforts with MINIVISION research group, which is the best result ever reported. Pls stay tuned for the solution paper. [22/01/2015]
Dr. Shuicheng YAN is entitled "Dean's Chair" associate professor. [14/01/2015]
Canyi LU won the "2014 Microsoft Research Asia fellowship award" founded on his series of research papers on "Block-diagonality Pursuit". [27/09/2014]
NUS-team is winner in Large Scale Visual Recognition Challenge 2014. It is great to see our "Network in Network" and "Deeper network in network" from Google winning detection and classification tasks respectively!
Dr. Yan is named IAPR Fellow for the contributions to computer vision and pattern recognition. [27/08/2014]
Dr. Yan is listed as ISI highly-cited Researcher 2014 in http://highlycited.com/. Highly Cited Researchers 2014 represents some of world's leading scientific minds. Over three thousand researchers earned the distinction by writing the greatest numbers of reports officially designated by Essential Science Indicators? as Highly Cited Papers -- ranking among the top 1% most cited for their subject field and year of publication, earning them the mark of exceptional impact. [11/06/2014]
Our group just released the new result on VOC2012 classification task,the performance is improved from best ever reported 0.828 to 0.930. See the PASCAL VOC leaderboard and the manuscript, CNN: Single-label to Multi-label. [09/06/2014]
Our (former) lab member, XU Zhongwen, the PhD student of Dr. Yi Yang, won the MSR-Bing Image Retrieval Challenge, associated with ICME 2014. [25/04/2014]
One paper from Bingbing Ni, "Order Preserving Sparse Coding", is accepted to be published in TPAMI.
One paper from Hairong LIU, "Dense Subgraph Partition of Positive Hypergraph", is accepted to be published in TPAMI.
One paper from Qiang CHEN, "Contextualizing Object Detection and Classification" is accepted to be published in TPAMI.
One paper from Jianqiu JI, "Batch-Orthogonal Locality-Sensitive Hashing for Angular Similarity", is accepted to be published in TPAMI.
One paper from Prof. HU Weiming's team and Qiang CHEN, "Bin Ratio-Based Histogram Distances and Their Application to Image Classification", is accepted to be published in TPAMI.
One paper from Zilei Wang and Jiashi Feng, "Collaborative Linear Coding for Robust Image Classification", is accepted to be published in IJCV.
One paper from Zhigang Ma, "Harnessing Lab Knowledge for Real-?world Action Recognition", is accepted to be published in IJCV.
One paper from Bingbing Ni, "Pose Adaptive Motion Feature Pooling for Human Action Analysis", is accepted to be published in IJCV.
Group member, Mr. LIU Luoqi, received the Best Student Paper Award (Golden Prize) from Pattern Recognition and Machine Intelligence Association (PREMIA) for the paper "Wow! You Are So Beautiful Today! in ACM . [14/02/2014]
NUS-team is the runner-up prize winner in Large Scale Visual Recognition Challenge 2013. The NUS-team performance is 12.95% error rate for top-5 results! Enjoy the by-product from this competition (though not used in the competition), our new Deep Learning architecture "Network in Network"; Thank Prof. Anthony Tung for the kind help of supporting us extra GPUs! [12/07/2013]
Congratulate the two awards won in ACM MM13 at Barcelona, Spain!
The best student paper award is for Hanwang Zhang, the student of Prof. Chua Tat-Seng.
S. Gong, M. Cristani, S. Yan, C.C. Loy (Eds.). "Person Re-Identification", 468 pages, Springer, December 2013.
We would like to acknowledge and congratulate you for the selection of "Visual Classification With Multitask Joint Sparse Representation" as a notable paper in computing in 2012. "Computing Reviews; Best of 2012 list" consists of book and article nominations from our reviewers, CR category editors, the editors in chief of journals we cover, and others in the computing community." -- From Computing reviews, a collaboration between ThinkLoud and the ACM. [05/28/2013]
Dr. Yan is co-chairing the Pacific-Rim Conference on Multimedia (PCM), Dec. 13-16, 2013, in Nanjing, China. The submission deadline is approaching! [05/06/2013]
Dr. Shuicheng Yan was promoted to associate professor with tenure with retrospective effect from 1 January 2013. [03/18/2013]
One paper from Xiaotong Yuan, "Forward Basis Selection for Pursuing Sparse Representations Over a Dictionary", is accepted to be published in TPAMI.
One paper from Guangcan Liu, "Robust Recovery of Subspace Structures by Low-?Rank Representation", is accepted to be published in TPAMI.
One paper from Hairong Liu, "Fast Detection of Dense Subgraph with Iterative Shrinking and Expansion", is accepted to be published in TPAMI.
Congratulate our former member Dr. Ni Bingbing (now research scientist in ADSC) and current members Tan Jun and Dong Jian for winning the "Human activities recognition and localization competition" contest in ICPR 2012!
Congratulate Tam V. Nguyen for being honored the 2nd prize winner of the "Kitchen Scene Context based Gesture Recognition" contest in ICPR 2012![12/30/2012]
Congratulate Mr. CHEN Qiang for his great work when doing intern at IBM! The joint team with Carnegie Mellon University and IBM T.J. Watson achieved No. 1 place in 4 out of 7 retrospective event detection tasks. [12/30/2012]
Graymatics, a startup company Dr. YAN Shuicheng is providing consultation on intelligent online advertisement, is featured on MIT Tech Review "Where Did You Get That Cool Shirt? This Software Knows" [10/09/2012]
Our joint team with Panasonic Singapore Laboratories (PSL) is the winner for the first time and also the last time of the object segmentation task in PASCAL VOC2012 Challenge! We achieved No.1 place for 12 classes from all 21 object classes!
Deep condolence to the great organizer Prof. Mark Everingham! [10/12/2012]
Our joint team with Panasonic Singapore Laboratories (PSL) and Sun Yat-sen University is the winner of the object classification task in PASCAL VOC2012 Challenge! We achieved No.1 place for all 20 object classes! From 2010 to 2012, the best performance is improved step-by-step as 0.73 - 0.78 - 0.82. [10/12/2012]
We have eight full papers, one short paper and three demos accepted to ACM MM 2012. Among them, we focused on clothes and beauty sense: 1) "Hi, Magic Closet, Tell Me What to Wear!", full paper by Si Liu and Jiashi Feng, and 2) "Sense Beauty via Face, Dressing, and/or Voice", full paper by NGUYEN VAN Tam and Si Liu. [08/08/2012]
Five papers plus one demo are accepted to ECCV12. The demo is Android based system "Object-Layout-Aware Image Retrieval for Personal Album Management", on how to intelligently use object layout for image search based on imperfect object detection results. [08/01/2012]
One paper from Guangcan Liu, "Active Subspace: Towards Scalable Low-Rank Learning" is accepted to be published in Neural Computation. [07/01/2012]
Dr. YAN Shuicheng is conferred the NUS Young Researcher Award, 2012 for outstanding contributions in multimedia, computer vision, and machine learning. [05/08/2012]
Seven papers are accepted to CVPR12, with one oral paper from Si LIU and Zheng SONG, "Street-to-Shop: Cross-Scenario Clothing Retrieval via Parts Alignment". [03/27/2012]
Dr. Yan is co-chairing the 19th International Conference on Multimedia Modeling, Jan. 07-09, 2013, in Huangshan, China. [02/20/2012]
Two papers from Xiaotong Yuan and Guangcan Liu are accepted to International Conference on Artificial Intelligence and Statistics (AISTATS), 2012. [02/20/2012]
Group member, Mr. CHEN Qiang & Mr. SONG Zheng (joint winner), received the Best Student Paper Award (Golden Prize) from Pattern Recognition and Machine Intelligence Association, Singapore (PREMIA) for their paper "Contextualizing Object Detection and Classification"in CVPR11, which is also the winner algorithm in PASCAL VOC 2010. [02/18/2012]
The paper "Image Re-emotionalizing" of Mengdi Xu and Bingbing Ni won the Best Paper Award in Pacific-Rim Conference on Multimedia (PCM), 2011. [12/24/2011]
Discovery.news reported NGUYEN VAN TAM's work "Seeing Your Weight – An application in targeted advertisement" shown in ICCV11 as a demo. [12/10/2011]
Physorg.com reported Bingbing NI and Zheng SONG's work "Web Image and Video Mining towards Universal and Robust Age Estimator" published in IEEE Transactions on Multimedia, 2011. [12/05/2011]
Dr. YAN Shuicheng is conferred the Singapore Young Scientist Award, 2011 for his research on computer vision, multimedia and machine learning. [11/12/2011]
In HHME11, Dr. YAN Shuicheng gave a keynote to introduce the new research direction on how multimedia and game may benefit from each other, Multimedia vs. Game. [09/17/2011]
One paper from Xiaotong Yuan, "Non-degenerate Piecewise Linear Systems: A Finite Newton Algorithm and Applications in Machine Learning", is accepted to be published in Neural Computation.
One paper from Hairong Liu, "Dense Neighborhoods on Affinity Graph", is accepted to be published in International Journal of Computer Vision. [09/08/2011]
Dr. YAN Shuicheng won the 2010 Young Faculty Research Award. [08/11/2011]
We have six papers and one demo accepted to ACM MM 2011, with one full paper on Multi-Semantic Image Annotation. [08/08/2011]
We have six papers accepted to ICCV 2011, with one ORAL presentation from SONG Zheng and NI Bingbing "Learning Universal Multi-view Age Estimator by Video Contexts". [07/23/2011]
Dr. YAN Shuicheng won the 2011 Best Associate Editor (BAE) Award of IEEE Transactions on Circuits and Systems for Video Technology(TCSVT). [04/29/2011]
Two papers on feature selection and block-diagonal regularization are accepted to AAAI11. [04/28/2011]
One paper from Xiaotong Yuan is accepted to International Conference on Artificial Intelligence and Statistics (AISTATS). [03/07/2011]
Four papers are accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2011, including one oral presentation. [03/07/2011]
Group member, Mr. CHENG Bin, received the Best Student Paper Award (second prize) from Pattern Recognition and Machine Intelligence Association, Singapore(PREMIA) for his TIP-2010 paper on L1-graph. [02/17/2011]
In ACM MM Multimedia Grand Challenge 2010 we present the work "Learning to Photograph", guiding unprofessional photographer to capture professional photos based on the photo compositional rules learnt from online professional photo corpus. [09/12/2010]
The joint work from Prof. Chua's group and our group: Richang Hong, Meng Wang, Mengdi Xu, Shuicheng Yan, Tat-Seng Chua, "Dynamic Captioning: Video Accessibility Enhancement for Hearing Impairment" won the best paper award of ACM MM10 in Italy. [10/30/2010]
Congratulations to Bingbing NI for being selected as "IBM Watson Emerging Leaders in Multimedia and Signal Processing 2010". The awardees, selected from a large pool of nominees, were highlighted in a two-day workshop presenting the state-of-the-art and trend of multimedia and signal processing research. [10/20/2010]
Our team is the winner of the object classification task in PASCAL VOC2010 Challenge, achieving No.1 place for 12 object classes among totally 20 classes! Also NUS team won the honorable mention prize in the object detection task! [09/12/2010]
Congratulations to Hairong Liu for his paper "Robust Clustering as Ensembles of Affinity Relations" accepted to NIPS 2010. [09/02/2010]
Group's paper "Automated Assembly of Shredded Pieces from Multiple Photos" from Shengjiao Cao and Hairong Liu won the Best Paper Award in IEEE International Conference on Multimedia Expo, 2010.
(associated love story. [07/21/2010]
We have six full papers accepted to ACM MM 2010. [07/15/2010]
LV group members witnessed the most precious moment of Bingbing and Xiuxiu! [04/26/2010]
We have seven papers accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2010, including two oral presentations. [03/01/2010]
This demo offers unique functionalities over the traditional advertising systems, (e.g., bill board, digital signage or multimedia advertising machines), and the style as well as content of the displayed media can change adaptively in real time according to the statistics of the age and gender of the audience.
(Presentation and Demo) [12/01/2009]
Our group member Bing-kun BAO's paper "Hidden-Concept Driven Image Decomposition Towards Semi-Supervised Multi-Label Image Annotation" won the Best Paper Award in International Conference on Internet Multimedia Computing and Service, 2009. [28/11/2009]
We have three full papers accepted to ACM MM 2009. [07/01/2009]
We have two collaborated ORAL papers accepted to IEEE International Conference on Computer Vision. [06/20/2009]
We have six papers accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR), including one for oral presentation. [02/26/2009]
We have one paper accepted to International Conference on Artificial Intelligence and Statistics (AISTATS). [01/09/2009]
Our group member, Mr. Ni Bingbing, received the Best Student Paper Award from Pattern Recognition and Machine Intelligence Association, Singapore (PREMIA) for his ICDM2008 paper. [02/07/2009]
Yan joined NUS/ECE on Dec. 21, 2007 and our first group member Mr. Bingbing Ni appeared on Dec. 22, 2007. Our research group website kicks off officially today. [12/10/2008]