|
This webpage shows the R&D works that I have done from 2006 to 2008. I'm still in the process of compiling these works into a reel, so I'm placing these individual works here into a single webpage for you to browse through quickly first. More details of these works can be found in other pages of this website if you are interested. Skills and experience in research
Fats Simulation (2008)
Created a new paradigm for fats simulation using a novel technique. More details to be available soon. Simplified Muscle Dyn. for Appealing Real-Time Skin Deformation (2007)
We propose significant simplifications in muscle modeling and simulation to facilitate real-time anatomical skin deformation for full-body articulated characters. The muscle shape is a function of an animated quadratic Bezier action curve and control rings derived from it. The action curve is uniformly sampled to derive control rings driven by a scaled sinusoidal equation to model fusiform shapes. A single spring is attached between the central control point and the midpoint vector between the extreme control points of the Bezier action curve. Care is taken to stabilize local coordinates for each muscle vertex to enable glitch-free skin deformation. The character’s polygonal mesh is smooth-skinned using a two-layered approach: first to the joints, and then to the muscle vertices. Lastly we show how different prominent muscles can be reasonably approximated with the proposed fusiform model. A typical 4000-vertex character skinned with sixty four 72-vertex muscles is able to run on an average CPU at 60-80 fps. Intuitive tools have been developed to aid the process of muscle creation, muscle shaping, dynamics simulation, as well as assignment of skin weights for joints and muscles. Our main contribution is the simplified dynamics driven curved action-axis, which enables economical and expressive muscle animation. This opens up new levels of realism for real-time characters in games and interactive media. Video showing our muscle system running in real-time, with and without muscle dynamics for comparison Using this technique, we were also able to create characters of different physiques on-the-fly by dynamically changing the muscle attributes.
Face Image Relighting (2007) In this project, we are trying to relight a face in a 2D image, given only a single image as reference. A 3D face model is obtained first by using a morphable face model. The matching/optimization then starts to find the best weights for the morphing so that the 3D morphed face matches the underlying image as best as possible. Finally, we relight the synthesized 3D face model and composite it back onto the 2d image to obtain the results. Video showing the matching process and results The technique was tested on two faces of different gender and different head facing angle. The results obtained were pretty good, given the fact that only 5 eigenvectors were used.
|
|