Just after Morishima-lab. was founded in 1987, topics about “face analysis and synthesis” formed a basic streams which was originated at “Model-based Image Coding” and the ancestor of this team produced “Future Cast System” and “Dive Into the Movie” as a pioneer of an innovative entertainment system. Recently, this team is challenging the wide variety of research topics about analysis and synthesis of facial expression, lip synchronization, face aging simulation, modeling of personalized face attribute and gaze simulation.
Especially, face texture reconstruction based on a patch tiling succeeds to produce a photo-realistic face animation especially including a 3D inner mouth expression using a data-driven approach. This method also contributes to generate a hyper resolution or de-blurred face image to improve an averaging or interpolated face image quality. “Detail-ization” has succeeds to produce a very photo-realistic speech production animation in which audience cannot find the difference between a CG generated face or a video captured face. “Character Transfer” is also a key technology to produce a very impressive face deformation and retargeting technique by keeping an original identity of characters as well as an animator’s hand crafted work with much fewer key faces than normal blend-shape animation. This team is also contributing to make a photo-realistic texture automatically from a hand written sketch or caricature, a photo-realistic aging simulation based on a skin quality considering personal wrinkles estimated from the location of smile wrinkles, a natural control method of gaze and head angle in a game character considering the location of a saliency of scenes and etc.
Face analysis and synthesis technique are basis of our teams and contributing to all of the topics in Morishima-lab. This team is financially supported by Ministry of Education, Culture, Sports, Science and Technology-Japan