个人简述:

2017年-至今 西北工业大学电子信息学院 教授。2012年 IEEE CVPR 最佳论文奖（Best Paper Award） (大陆高校首次获得) 2014年 DICTA DSTO 图像处理最佳基础贡献论文奖 2017年 IEEE CVPR 非刚性结构与运动恢复挑战赛最佳算法奖2017年 DICTA 最佳学生论文奖2017年 APSIPA ASC最佳机器学习/深度学习论文奖

科研工作:

2018Liyuan Pan, Yuchao Dai, MiaomiaoLiu, Fatih Porikli. Depth Map Completion by Jointly Exploiting Blurry ColorImages and Sparse Depth Maps. WACV 2018 Accepted.2017Liu Liu, Hongdong Li, Yuchao Dai and Quan Pan. Robust andEfficient Relative Pose with a Multi-camera System for Autonomous Driving inHighly Dynamic Environments. IEEE Transactions on Intelligent TransportationSystems (T-ITS). In Press.Suryansh Kumar, YuchaoDai, Hongdong Li. Monocular Dense 3D Reconstruction of a Complex DynamicScene from Two Perspective Frames. ICCV 2017.LiuLiu, Hongdong Li, YuchaoDai. Efficient Global 2D-3D Matching for Camera Localization in aLarge-Scale 3D Map. ICCV 2017.Pan Ji, Hongdong Li, Yuchao Dai, Ian Reid. "MaximizingRigidity" Revisisted: A Convex Programming Approach for Generic 3D ShapeReconstruction from Multiple Perspective Views. ICCV 2017.Liyuan Pan, Yuchao Dai, Miaomiao Liu, Fatih Porikli.Simultaneous Stereo Video Deblurring and Scene Flow Estimation. CVPR 2017.Suryansh Kumar, Yuchao Dai, Hongdong Li.Spatio-Temporal Union of Subspaces for Multi-body Non-rigidStructure-from-Motion. PatternRecognition. 71(11): 428-443, 2017.Dingfu Zhou, Vincent Fremont, Benjamin Quost, Yuchao Dai,Hongdong Li. Moving Object Detection and Segmentation in Urban Environmentsfrom a Moving Platform. Imageand Vision Computing specialissue on Automotive Vision. In Press.Jing Zhang, Yuchao Dai, Bo Li, Mingyi He.Attention to the Scale: Deep Multi-Scale Salient Object Detection. DICTA 2017.(Best Student Paper Award).Yuchao Dai, Huizhong Deng, Mingyi He.Dense Non-Rigid Structure-from-Motion Made Easy – A Spatial-Temporal SmoothnessBased Solution. ICIP 2017.Jing Zhang, Bo Li, Yuchao Dai, Fatih Polikli, Mingyi He.Integrated Deep and Shallow networks for salient object detection. ICIP 2017, Oral Presentation.Zhipeng Xiao, Hongdong Li, Dingfu Zhou, Yuchao Dai, BinDai. Accurate Extrinsic Calibration between Monocular Camera and Sparse 3DLidar Points without Markers. IEEE Intelligent Vehicles Symposium (IV 2017).Jing Zhang, Yuchao Dai, Fatih Porikli. Deep SalientObject Detection by Integrating Multi-level Cues. WACV 2017.Jing Zhang, Yuchao Dai, Fatih Polikli,Mingyi He. Multi-Scale Salient Object Detection with Pyramid Spatial Pooling. APSIPA ASC 2017. (Best Machine Learning/Deep Learning Paper Award).2016Suryansh Kumar, Yuchao Dai, Hongdong Li. Multi-body Non-rigidStructure-from-Motion. 3DV 2016.Xibin Song, Yuchao Dai, Xueying Qin. Deep DepthSuper-Resolution : Learning Depth Super-Resolution using Deep ConvolutionalNeural Network. ACCV 2016.Yuchao Dai, Hongdong Li, Laurent Kneip.Rolling Shutter Camera Relative Pose - Generalized Epipolar Geometry. CVPR 2016.Jiaolong Yang, Hongdong Li, Yuchao Dai, RobbyTan. Robust Optical Flow Estimation of Double-Layer Images under Transparencyor Reflection. CVPR 2016.Dingfu Zhou, Yuchao Dai,Hongdong Li. Reliable Scale Estimation and Correction for Monocular VisualOdometry. IEEE Intelligent Vehicles Symposium (IV 2016).Huizhong Deng, Yuchao Dai. Pushingthe limit of Non-rigid Structure-from-Motion by Shape Clustering. ICASSP 2016.Yuchao Dai, Hongdong Li. SimultaneousCorrespondences Estimation and Non-rigid Structure Reconstruction. DICTA 2016.2015Bo Li, Chunhua Shen, Yuchao Dai,Anton van den Hengel, Mingyi He Depth and Surface Normal Estimation FromMonocular Images Using Regression on Deep Features and Hierarchical CRFs. CVPR 2015, 1119-1127Jiayan Qiu, Yuchao Dai,Yuhang Zhang, Jose M. Alvarez Hierarchical Aggregation Based Deep AgingFeature for Age Prediction. DICTA2015: 1-5.2014Yuchao Dai, Hongdong Li, Mingyi He. “Asimple prior-free method for nonrigid structure from motionfactorization”. InternationalJournal of Computer Vision (IJCV). 107(2) 101-122, 2014. (ISI Impact Factor=8.222) (IJCV is top journal in ArtificialIntelligence and Image Processing.) [Code]Yuchao Dai, Hongdong Li. Rankminimization or Nuclear norm minimization: Are we solving the right problem? DICTA 2014. (The DSTO Best Fundamental Contribution to Image Processing Paper Prize Awarded by Defence Science andTechnology Organisation (DSTO).)Pan Ji, Hongdong Li, Mathieu Salzmann, Yuchao Dai: RobustMotion Segmentation with Unknown Correspondences. ECCV (6) 2014: 204-219Wei Zhou Yuchao Dai Renjie He, "Efficient depthestimation from single image," Signal and Information Processing (ChinaSIP), 2014 IEEE China Summit &International Conference on, vol., no., pp.296,300, 9-13 July 2014Bo Li Yuchao Dai Mingyi He van den Hengel, A.,"A relaxation method to articulated trajectory reconstruction frommonocular image sequence," Signal and Information Processing (ChinaSIP), 2014 IEEE China Summit &International Conference on, vol., no., pp.389,393, 9-13 July 2014Jing Zhang Mingyi He Yuchao Dai,"Modified affinity propagation clustering," Signal andInformation Processing (ChinaSIP),2014 IEEE China Summit & International Conference on, vol., no.,pp.505,509, 9-13 July 20142013Yuchao Dai, Hongdong Li, Mingyi He. “ProjectiveMulti-view Structure and Motion from Element-wise Factorization”. IEEE Transactions onPattern Analysis and Machine Intelligence (TPAMI), vol.35, no.9, pp.2238-2251, Sept.2013. (ISI Impact Factor=8.329) (TPAMI is top journal in ArtificialIntelligence and Image Processing.) [Code] (Both the SDPVersion and the ADMM Version)Richard Hartley, Jochen Trumpf, Yuchao Dai, and Hongdong Li. “Rotation Averaging”. InternationalJournal of Computer Vision (IJCV).July 2013, Volume 103, Issue 3, pp 267-305 (ISI Impact Factor=8.222) (IJCV is top journal in ArtificialIntelligence and Image Processing.)Jae-hak Kim*, Yuchao Dai*, Hongdong Li, Xin Du and Jonghyuk Kim, Multi-View3D Reconstruction from Uncalibrated Radially-Symmetric Cameras. InternationalConference on Computer Vision, ICCV2013 (* Equal Contribution).Jiaolong Yang, Yuchao Dai, Hongdong Li, Henry Gardner, and YundeJia. "Single-Shot Extrinsic Calibration ofa Generically Configured RGB-D Camera Rig from Scene Constraints". ISMAR 2013.(Full paper, oral presentation, acceptance rate is 25.2%) .2012Yuchao Dai, Hongdong Li, Mingyi He. CVPR2012. (Oral) [Acceptance rate is 2.5% , i.e.48 papers accepted as oral, out of 1933 submissions] Winner of CVPR Best Paper Award 2012. Video Code (Matlab source code is freelyavailable for academic users.) [Code]Yuchao Dai, Mingyi He, Hongdong Li, Bo Li. GenericSmoothness Constrained 3D Trajectory Reconstruction of Moving Object fromMonocular Video. China Science:Information Science. 2012, 42(8):988-1006.2011Yuchao Dai, Mingyi He, Hongdong Li,Richard Hartley, Factorization-based Structure-and-motion Computation forGeneralized Camera Model. IEEE International Conference on SignalProcessing, Communications and Computing, 2011 (ICSPCC 2011), 1-6.2010Yuchao Dai, Hongdong Li, Mingyi He. Element-wise Factorization for N-ViewProjective Reconstruction. ECCV 2010. [AcceptRatio: 322/1174=27.4%] (Awarded a student grant). 396-409.Richard Hartley, Jochen Trumpf, Yuchao Dai, Rotation averaging and weak convexity. In Proceedings ofthe 19th International Symposium on Mathematical Theory of Networks and Systems(MTNS), pages 2435-2442, 2010.2009Yuchao Dai, Jochen Trumpf, Hongdong Li,Nick Barnes, Richad Hartley, Rotation Averaging with Applicationto Camera-Rig Calibration, In Asian Conference on Computer Vision(ACCV 2009) [AcceptRatio:？175/670=27.1%],September 2009, Xian, ChinaYuchao Dai, Hongdong Li, Mingyi He,Chunhua Shen, Smooth Approximation for L-Infinity Norm, In: InternationalConference on Digital Image Computing - Techniques and Applications (DICTA 2009), December 2009. Melbourne,Australia. Yuchao Dai, Mingyi He, Hongdong Li, TwoEfficient Algorithms for Outlier Removal in Multi-view Geometry UsingL-Infinity Norm, In International Conference on Image and Graphics (ICIG 2009), Xi'an, China2008Mingyi He Yuchao Dai JingZhang Image registration by integrating similarityand epipolor constraints IEEE conference on IndustrialElectronics and Applications(ICIEA 2008), June 2008, Singapore. Page(s):1870-1874.Yuchao Dai, Mingyi He, Jing Zhang, TaoGao, Stereo Matching based on RBF Neutral Network, Application Research of Computers,2008(11), In Chinese

教育背景:

200109-200507西北工业大学电子信息工程工学学士；200509-200804西北工业大学信号与信息处理工学硕士；200709-201205西北工业大学信号与信息处理工学博士；200810-200910澳大利亚国立大学信息科学与工程访问博士生。