3D shape measurement using image-matching-based techniques

Image matching involves detecting the same points in two or multiple images that are captured from different viewpoints, at different time, and/or by different cameras. This paper presents using the image-matching-based techniques to carry out the static and dynamic 3D shape measurements. The process contains two crucial steps: (1) calibrate the cameras to get the intrinsic and extrinsic parameters; (2) perform matching of pixel points to detect the location disparities of the same physical points in the involved two or multiple images. A number of experiments are shown to demonstrate the applications to 3D shape, deformation, motion and vibration measurements.