An Improved Algorithm Robust to Illumination Variations for Reconstructing Point Cloud Models from Images

Reconstructing 3D point cloud models from image sequences tends to be impacted by illumination variations and textureless cases in images, resulting missing parts or uneven distribution of retrieved points. To improve the reconstructing completeness, this work proposes an enhanced similarity metric which is robust among images during dense diffusions push seed-and-expand scheme a further extent. This integrates zero-mean normalized cross-correlation coefficient that texture information respectively weakens influence cases. Incorporated with disparity gradient confidence constraints, candidate features are diffused their neighborhoods for points recovering. We illustrate two-phase results multiple datasets evaluate robustness proposed algorithm variations. Experiments show ours recovers 10.0% more points, on average, than comparing methods varying scenarios achieves better completeness comparative accuracy.