Robotics and Computer Vision Lab

Publications

조회 수 1850 댓글 0
Extra Form
저 자 Jiyoung Jung, Joon-Young Lee, In So Kweon
학 회 20th Korea-Japan Joint Workshop on Frontiers of Computer Vision
Notes This research was supported by the MOTIE (The Ministry of Trade, Industry and Energy), Korea, under the Human Resources Development Program for Convergence Robot Specialists support program supervised by the NIPA (National IT Industry Promotion Agency) (H1502-13-1001)
논문일시(Year) 2014
논문일시(Month) 02
A time-of-flight camera provides depth maps of the scene at video frame rate. However, their depth measurements are severely influenced by random noise and systematic bias. Previous approaches on depth denoising are usually variants of adaptive joint bilateral filtering with the help of a color image of the same scene. In this paper, we access to the raw range measurements of the ToF sensor instead of the transformed depth values, and we acquire range error profi le for each pixel along the range measurement by capturing a planar scene at di fferent distances. We correct the range bias using plane fi tting and then the remaining noise can be assumed to follow a zero-mean Gaussian distribution with variance according to the pixel location and the range measurement. Since the whole process is done beforehand leaving variance information, any kind of depth denoising algorithm assuming zero-mean Gaussian noise can perform well with our noise estimation.

등록된 글이 없습니다.

Board Pagination Prev 1 Next
/ 1