| Abstrakt: |
Three dimensional symmetry plane detection is a hot research topic in the field of computer vision. When detecting the symmetry plane, the integrity of the three-dimensional point cloud is often ignored, and it is often defaulted to be complete and absolutely symmetrical, which makes the mirror key points relatively easy to be found.This proposes a method for 3D symmetry plane detection based on 2D image and transformation. In detail, the proposed method firstly detects the mirror key points in a single 2D image of the target, then selects the corresponding points in the 2D image and the 3D point cloud to construct a transformation matrix, and finally obtains the mirror key points in the 3D point cloud based on the 2D mirror key points and the transformation matrix to detect the symmetry plane. Experimental evaluations are performed on both synthetic and real point cloud datasets. The results show that the proposed approach is effective for the complete point cloud as well as the incomplete point cloud. Compared with the other two methods, it is proved that the symmetry plane detected by the proposed method is more accurate. The experimental results show that the symmetry plane detected by the proposed method is more accurate than the other two comparison methods. Then the experimental results on incomplete point clouds show that the proposed method could detect symmetry plane effective. [ABSTRACT FROM AUTHOR] |
Full text from SpringerLink