zhuangqirong
commented
1 year ago When using train_net.py to train, an error occurs when performing eval： cv2.error:resize.cpp:4062:erroe:(-215:Assert failed) !ssize.empty() in function 'resize'
Locating errors 90-93 in dtu/enerf.py tar_dpt = data_utils.read_pfm(scene_info['dpt_paths'][tar_view])[0].astype(np.float32) tar_dpt0(128, 160) tar_dpt = cv2.resize(tar_dpt, None, fx=0.5, fy=0.5, interpolation=cv2.INTER_NEAREST) tar_dpt1(64, 80) tar_dpt = tar_dpt[44:556, 80:720] tar_dpt2(20, 0) tar_mask = (tar_dpt > 0.).astype(np.uint8) tar_mask(20, 0)
After spliting, tar_mask is empty how to solve it？
How do you solve this problem?
When using train_net.py to train, an error occurs when performing eval： cv2.error:resize.cpp:4062:erroe:(-215:Assert failed) !ssize.empty() in function 'resize'
Locating errors 90-93 in dtu/enerf.py tar_dpt = data_utils.read_pfm(scene_info['dpt_paths'][tar_view])[0].astype(np.float32) tar_dpt0(128, 160) tar_dpt = cv2.resize(tar_dpt, None, fx=0.5, fy=0.5, interpolation=cv2.INTER_NEAREST) tar_dpt1(64, 80) tar_dpt = tar_dpt[44:556, 80:720] tar_dpt2(20, 0) tar_mask = (tar_dpt > 0.).astype(np.uint8) tar_mask(20, 0)
After spliting, tar_mask is empty how to solve it？