From synthesis to analysis: fitting human animation models to image data

Abstract

We show that we can effectively fit complex animation models to noisy image data. Our approach is based on robust least-squares adjustment and takes advantage of three complementary sources of information: stereo data, silhouette edges and 2-D feature points. We take stereo to be our main information source and use the other two whenever available.In this way, complete head models-including ears and hair-can be acquired with a cheap and entirely passive sensor, such as an ordinary video camera. The motion parameters of limbs can be similarly captured. They can then be fed to existing animation software to produce synthetic sequences.

Animated Heads from Ordinary Images: A Least-Squares Approach

“…Here the motion of the head was recovered using an automated bundle adjustment algorithm [9,10]. Following the complete procedure outlined in Section 2, we first recovered the face as before.…”

The Role of Virtual Humans in Virtual Environment Technology and Interfaces

“…For head, although t is possible to create them using an interactive sculpting tool, the best way is to reconstruct them from reality. Three methods have been used for this: 1) Reconstruction from 2D photos [3] 2) Reconstruction from a video sequence [4] 3) Construction based on the laser technology The methods could be used for body modelling, but the main problem is still with the body deformations which has been addressed by many researchers, but is still not 100% solved.…”

The Simulation and Optimization of Job-Shop Equipment Based on JACK

“…Currently,physical body models usually include anthropometry model, biomechanical model and user interface model [1]. JACK software is a mature software for human simulation model and ergonomics evaluation, which includes basic anthropometric data, the degree of fatigue and other medical and physiological parameters [2][3][4]. It can be used to study the virtual human motion analysis and virtual environments [5] .…”