Deep-Learning-Based Multiview RGBD Sensor System for 3-D Face Point Cloud Registration

The 3-D face analysis is now a hot research topic in computer vision. The acquisition of 3-D faces requires acquiring multiview face and obtaining them by point cloud registration. In this letter, a deep-learning method is introduced in the multiview RGB-depth (RGBD) depth sensor system to acquire matching point pairs of image data. Thus, 3-D spatial point pairs are obtained. Then, the initial poses between the point clouds of multiview faces are obtained by rigid body transformation estimation, and the coarse registration of the point clouds is completed. On this basis, for fine registration of multiview 3-D face point clouds using the iterative closest point algorithm, this letter proposes a method to convert partially overlapping point cloud registration into subregion point cloud registration through facial region positioning. In this letter, the maximum value of root mean square error (RMSE) for the experimental results of point cloud coarse registration is 2.26 mm with the mean and standard deviation of 1.18 ± 0.33 mm; the maximum value of RMSE for the experimental results of point cloud fine registration is 1.47 mm with the mean and standard deviation of 0.91 ± 0.14 mm. It is confirmed that the proposed method is very robust and effective for the registration of multiview 3-D face point clouds.