Rigid optimizer option class.
__init__(self, maximum_iteration=0, maximum_allowable_depth=2.5, depth_threshold_for_visibility_check=0.03, depth_threshold_for_discontinuity_check=0.1, half_dilation_kernel_size_for_discontinuity_map=3, image_boundary_margin=10, invisible_vertex_color_knn=3, debug_output_dir='')¶
maximum_iteration (int, optional, default=0) – int: (Default
300) Number of iterations for optimization steps.
maximum_allowable_depth (float, optional, default=2.5) – float: (Default
2.5) Parameter to check the visibility of a point. Points with depth larger than
maximum_allowable_depthin a RGB-D will be marked as invisible for the camera producing that RGB-D image. Select a proper value to include necessary points while ignoring unwanted points such as the background.
depth_threshold_for_visibility_check (float, optional, default=0.03) – float: (Default
0.03) Parameter for point visibility check. When the difference of a point’s depth value in the RGB-D image and the point’s depth value in the 3D mesh is greater than
depth_threshold_for_visibility_check, the point is marked as invisible to the camera producing the RGB-D image.
depth_threshold_for_discontinuity_check (float, optional, default=0.1) – float: (Default
0.1) Parameter to check the visibility of a point. It’s often desirable to ignore points where there is an abrupt change in depth value. First the depth gradient image is computed, points are considered to be invisible if the depth gradient magnitude is larger than
half_dilation_kernel_size_for_discontinuity_map (int, optional, default=3) – int: (Default
3) Parameter to check the visibility of a point. Related to
depth_threshold_for_discontinuity_check, when boundary points are detected, dilation is performed to ignore points near the object boundary.
half_dilation_kernel_size_for_discontinuity_mapspecifies the half-kernel size for the dilation applied on the visibility mask image.
image_boundary_margin (int, optional, default=10) – int: (Default
10) If a projected 3D point onto a 2D image lies in the image border within
image_boundary_margin, the 3D point is considered invisible from the camera producing the image. This parmeter is not used for visibility check, but used when computing the final color assignment after color map optimization.
invisible_vertex_color_knn (int, optional, default=3) – int: (Default
3) If a vertex is invisible from all images, we assign the averaged color of the k nearest visible vertices to fill the invisible vertex. Set to
0to disable this feature and all invisible vertices will be black.
debug_output_dir (str, optional, default='') – If specified, the intermediate results will be stored in in the debug output dir. Existing files will be overwritten if the names are the same.