Dr Shaodi You

PhD Engineering, University of Tokyo. (2015)
Researcher / Honorary Lecturer
College of Engineering & Computer Science

Areas of expertise

  • Computer Vision 080104
  • Computer Graphics 080103
  • Image Processing 080106
  • Pattern Recognition And Data Mining 080109
  • Virtual Reality And Related Simulation 080111

Research interests

Physics based vision, image/video enhancement, prosthetic vision and machine learning.


Dr. Shaodi You is a senior research scientist at Data61, (Previously known as NICTA) Canberra Research Lab and an adjunct lecturer of College of Engineering and Computer Science at Australian National University (ANU). He receives his Ph.D. and M.E. degrees from The University of Tokyo, Japan in 2015 and 2012 and his bachelor's degree from Tsinghua University, P. R. China in 2009. His research interests are 1. physics based vision, 2. perception based vision and learning, and 3. 3D geometry.

Research Webpage: users.cecs.anu.edu.au/~shaodi.you/

Researcher's projects

My work is mainly focusing on three areas:
Perception Based Vision and Learning,
Physics Based Vision
3D geometry


Perception Based Vision and Learning


Perceptual based vision learning aiming to model and understanding human perception. And later propose computer vision and machine learning algorithms which incorporates such understanding. Specifically, I have been working on the topics: Prothetic vision, decolourization, saliency and manifold learning. (See research and projects)



Physics Based Vision


Liquid optics is extremely challenging because it is transparent and non-rigid. The transparency means its appearance is totally determined by its environment. The non-rigidity means its shape is also highly dependent on the environment. Specifically, my research focus on rain and water drops. (See research and projects)



Non-rigid 3D geometry


A few researches on modeling the non-rigidity of 3D objects. (See research and projects)

Current student projects

Project 1: Human Validation of RGB-D Saliency
Background and motivation: Recently, Image processing and computer vision are very popular research areas while saliency object detection and eye tracking are important in this space. The part which has a quality that thrusts itself into attention is called salient object in the picture and eye tracking is a process of measuring either the point of gaze or the motion of an eye relative to the head. Methods of salient object detection and eye tracking in a 2D scene are mature and developed which have been commercialized and used widely, but the similar techniques in 3D scene still needs exploration, with the growing imaged techniques like virtual realities (VR), it is becoming more and more important, so we will focus the saliency detection and eye tracking system in 3D scene.

Research topic and goal: The topic will be the methods of salient object detection and application of eye tracking system in 3D scene, the goal of the project is to design and build a system which can automatically and rapidly detect salient object in 3D scene.


Project 2: Physical Realistic Simulation for Automatic Driving in Bad Weather

Background and motivation: It reported that the annual rate of deaths and serious injuries in traffic accident is more than 1,200 in Australia, even though the situation has been improved in past year because of the development passive safety system. It is suggested that active safe system like automatic driving system is integrated in to cars is able to provide more safe and fast service. Automatic system will be tested under simulator before in practical test in the early stage of development. The realistic dynamic simulation environment enables a harmless representation of critical driving situation like extreme weather. Thus, the simulator plays essential role in early stage of development of automatic system. In this project, we are going to build driving simulator to provide excellent tool for development, testing, and research for automatic driving system.

Research topic and goal: The project is aimed to provide dynamic simulator environment in order to perform the automatic driving test in a virtual environment. The simulator allows the automatic system be tested under critical traffic situation and extreme driving maneuvers before being tested in realistic scenarios. The Various measure is implemented in the driving simulator in order to provide a reality-like environment for automatic driving. Because of ground truth, however, the measurement for fog and haze is inaccurate when they are implemented in the simulator. In that sense, physical models need to be built to achieve to achieve more realistic haze or fog simulation environment. Also, the physical model on haze and fog situation will be integrated into simulator.

Return to top

Updated:  01 December 2020 / Responsible Officer:  Director (Research Services Division) / Page Contact:  Researchers