Three Dimensional Super Resolution Reconstruction of Sparse Texture Image With Multi Depth Camera Calibration

The super resolution reconstruction method based on learning model is complex, the reconstructed image is fuzzy and the resolution is low. In order to solve this kind of problem, put forward the sparse texture image reconstruction algorithm of multi camera calibration depth, gives the detection data sequence of 3D reconstruction, image contour extraction method using blue screen, multi depth camera calibration by the Bundler open source software, obtain good calibration of multi view images. Using SIFT algorithm to extract the feature points in the image, the use of three-dimensional reconstruction algorithm based on PMVS images, sparse texture image reconstruction. The experimental results show that the proposed method can improve the resolution of the coefficient texture image, and has high performance and robustness.