Yuchao Dai


Letting the Computer See and Understand


July 2017, Best Algorithm Prize at CVPR 2017 Non-rigid Structure-from-Motion (NRSFM) Challenge!


July 2017, Three papers accepted by ICCV 2017.

-Monocular dense non-rigid 3D reconstruction;

-Maximizing rigidity for 3D reconstruciton;

Large scale camera localization.


May 2017, work on Spatio-Temporal Union of Sub-space Accepted by Pattern Recognition.


May 2017, Two Paper Accepted by ICIP 2017, one on Salient Object Detection and the other one on Non-rigid Structure-from-Motion.  

March 2017, work on Stereo Deblurring and Scene Flow Estimation Accepted by CVPR 2017


March 2017, work on Extrinbsic Calibration between Lidar and Camera Accepted by IV 2017


December 2016, work on Salient object detection accepted by WACV 2017.


September 2016, work on multi-body NRSFM acetped by 3DV 2016.


August 2016, work on depth superresolution by deep convolutional network accetped by ACCV 2016.


April 2016, work on robust scale estimation accepted by IV 2016


March 2016, two papers accepted by CVPR 2016.

- Rolling shutter relative pose

- Robust Optical Flow Estimation of Double-Layer Images under Transparency or Reflection


December 2015, work on NRSFM "Pushing the limit of Non-Rigid Structure-from-Motion by Shape Clustering "accepted by ICASSP 2016


March 2015, work on monocular depth estimation by using deep CNN was accepted by CVPR 2015.



Research School of Engineering

College of Engineering and Computer Science

The Australian National University

Ph.D  Email: yuchao.dai@anu.edu.au



I am currently an ARC DECRA Fellow with the Research School of Engineering at the Australian National University, Canberra. I received the B.E. degree, M.E degree and Ph.D. degree all in signal and information processing from Northwestern Polytechnical University, Xian, China, in 2005, 2008 and 2012, respectively. I was a visiting student at ANU from Oct. 2008 to Oct. 2009 with the support of the China Scholarship Council. My research interests include structure from motion, multi-view geometry, human-computer interaction, compressive sensing and optimization. I have published papers in both top-ranked journals and prestigious conferences such as IEEE TPAMI, IJCV, ICCV, CVPR and ECCV. I won the best paper award in IEEE CVPR 2012.

My recent work aims at developing dense non-rigid structure-from-motion methods, deep learning (depth estimation, saliency detection, super-resolution), low-level computer vision (image deblur, super-resolution, saliency detection), fine gesture recognition from RGB-Depth video cameras, with a view to support surgeons in their interaction with medical imagery.

Powered ByBy Google Sites