Intelligent Vision System research team.
A leader in advanced low cost computational 3D vision solutions.
Recent Advances in Online Computational Stereo Vision
We present a fully featured, web-based, online autostereogram creation system that allows a user to upload their own stereo images, generate depth data via computational stereo vision, and then turn this depth data into an autostereogram. The system can also perform the reverse process and extract depth data from a given autosteregram or generate anaglyphs from them.
Belief Propagation Based Stereo
The belief propagation based stereo approach approximates the minimum energy solution on graphical models such as Markov Chains, or Markov Random Field (MRF) of disparities. Our approach exploits a symmetric Cyclopean matching model, which accounts for visibility conditions, to construct epipolar profiles which are close to the human perception.
The use of Graphics Processing Units (GPUs)in computationally intensive applications is increasing due to recent frameworks that allow general purpose programming to be done on the specialized hardware of the GPU. This project summarizes the use of a GPU to improve the performance of a Smoothed Particle Hydrodynamics (SPH) program.
Real-time 3D hand tracking for 3D sculpting application
Although most of the existing 3D sculpting programs (e.g. Pixologic Z-brush, Autodesk Mudbox, etc.) can be controlled by special hand-controlled hardware such as a conventional or 3D mouse, or a tablet, it will be more convenient and natural to let users sculpt 3D models just by motions and poses of bare hands.
Developing intelligent vision system to detect, classify, or retrieve natural 3D objects under varying imaging conditions and operating environments is one of the most challenging artificial intelligence and computer vision problems having considerable practical interest. In particular, computer stereo vision can serve as an important tool in photogrammetry, cartography, civil engineering, mining, forensics, archaeology, biometrics, autonomous navigation, augmented reality, game industry, special movie effects, etc. Many applications is strong demand (e.g. security checks by face recognition, or terrain changes detection_ require fast and accurate responses. We are developing efficient solutions for various application problems that need real-time or near real-time modelling of close-range 3D environment.
Our current applications: Rapid prototyping of indoor and outdoor 3D scene and objects 3D modelling and animation of faces Web-based 3D computer vision 2D and 3D face recognition Real-time visualisation of 3D scenes Technology showcased: Robust stereo-matching under adverse lighting and contrast conditions. GPU programming with CUDA Multi-sensor data fusion techniques for scene and environment modelling Visualisation on 3D display 3D face modelling from depth maps acquired by stereo Integrated acquisition-to-visualisation system using stereo acquisition setup, powerful PC and 3D display output.