Locality preserving speaker clustering
Stephen M. Chu, Hao Tang, et al.
ICME 2009
Two results are presented in this paper. First, it is shown that in estimating three-dimensional motion of a rigid planar patch, the eight pure parameters used in [1] and [2] are uniquely determined from the image correspondences of four points, no three collinear, and can be estimated by solving a set of linear equations. The second result concerns the three-view problem. It is proved that given four image point correspondences in three perspective views of a planar patch undergoing general three-dimensional rigid body motion, the number of solutions for the motion parameters is one, as opposed to two [2] when only two perspective views are given. © 1984 IEEE
Stephen M. Chu, Hao Tang, et al.
ICME 2009
Shiyu Chang, Yang Zhang, et al.
NeurIPS 2017
Yang Fu, Yunchao Wei, et al.
ICCV 2019
Reimar K. Lenz, Roger Y. Tsai
IEEE Transactions on Pattern Analysis and Machine Intelligence