Problem with visualisotion. #18
Description
I use the "config.yaml" with params
'base_dir': "DATA" #The path of directory including img,pcd,out folders.
'image_res': (2736,5472) #image resolution for panoramic image only
'backend': 'opencv' #backend of detecting corners from the image, "matlab" or "opencv"
'output_img_with_dectected_corners': True #output the image with detected corners
'back_proj_corners': True #back-project the detected corners of the point cloud to the images
'camera_type': 'panoramic' #camera model: 'panoramic' or 'perspective'
'instrinsic_para': (166.87739756, 0., 311.29512622, 0., 334.91696616, 781.89612824, 0., 0.,1.)
intrinsic parameters are necessary for perspective cameras (1,2,3,4,5,6,7,8,9)→ [1,2,3, 4,5,6, 7,8,9]
Parameters for segmenting the point cloud
'file_name_digits': 4 #filename format
'LiDAR_type': 'vlp16_puck' # available choices 'vlp16_puck', 'hdl32', 'hdl64'
'laser_beams_num': 16 #The number of the laser beams of the Velodyne LiDAR
'jdc_thre_ratio': 8 #The ratio of adaptive threshold for clustering scanline
'agglomerative_cluster_th_ratio': 8 #The ratio of adaptive threshold for combining the scanline cluters
Parameters for detecting chessboard from segmentation result
'pattern_size': (8,8) #number of vertical and horizontal patterns
'grid_length': 0.035 #the length of one grid of the pattern [m]
'chessboard_detect_planar_PCA_ratio': 0.05 # The threshold of the planarity check of the potential chessboard with PCA
'marker_range_limit': 2 #The farthest range of the chessboard for filtering irrelative segments. Segments whose centroid to Velodyne farther that this threshold will not be considered as a potential chessboard.
Parameters for 3D corner detection from detected chessboard point cloud segment
'start_pattern_corner': 0 #The color of the pattern in the left down of the chessboard and 0 for black 1 for white
'intensity_col_ind': 6 # The index of intensity column (counting starts from 0). !!! Wrong number will cause the failure of corner detection in pcd files.
Settings for multi-processing
'multi_proc': False #Use multiple processing
'proc_num': 2 #The number of cores to use for multiple processing
'poses_num': 7 #The number of image-laserscan pairs
Ahd when I use the script:
from ILCC import utility
utility.vis_back_proj(ind=1, img_style="orig", pcd_style="dis", hide_occlussion_by_marker=False)
utility.vis_back_proj(ind=1, img_style="orig", pcd_style="dis", hide_occlussion_by_marker=True)
utility.vis_back_proj(ind=1, img_style="edge", pcd_style="intens", hide_occlussion_by_marker=True)
I have the following result:
What is the reason of this?