LIANG Xingang1,2
(1 Key Laboratory of Modern Teaching Technology, Ministry of Education, Xi′an 710062, Shaanxi, China;2 School of Computer Science, Shaanxi Normal University,Xi′an 710119, Shaanxi, China)
Abstract:
To reconstruction the 3D non-rigid from a un-calibration image sequence, a factorization method for 3D non-rigid reconstruction is presented. Firstly, based on the character that the image matrix composed by all the image points and depth factors is a low rank, the projective reconstruction which is different a transformation matrix with the real one is obtained. The transformation matrix can be linearly solved based on the constraints of the projective matrix, and the reconstruction can be upgrade to the Euclid one. The innovation of the method is that the solving of the transformation matrix is linear and all the images are treated uniformly. The experiments with both simulate and real data show that the method presented in the paper is efficient.
KeyWords:
non-rigid; 3D reconstruction; factorization
