How I get the occlusion part in the SCARED dataset?

[fromWRF]

As I can see in your data structure, there is an occlusion folder in your data, e.g. Scene Flow, MPI Sintel, KITTI 2015, MIDDLEBURY_2014, SCARED data. You give the answer about how to get the occlusion in the Scene Flow data, however, how to get the occlusion part in the SCARED data?
Thanks for your answer!!!

[Miaowei-HNU]

Can you find it？

[mli0603]

Hi @Miaowei-HNU

The occlusion is computed from the following steps

• The provided depth map are backprojected to both left and right cameras using the preprocessing scripts here
• The occlusion is computed using the script here

[Miaowei-HNU]

Thanks for your answer @mli0603

[Laqqq]

Hi! Thank you for providing this information, but in calculating the occlusion, how did you get the right disparity? As I follow the step but only get the left one. Thank you for clarification!

[MichaelWangGo]

Hi, thanks for providing the method, but the occ_mask_l is returned as True and False in each pixel:
occ_mask_l.shape (1024, 1280, 3)
occ_mask_l [[[ True False False]
[ True False False]
[ True False False]
...
[ True False False]
[ True False False]
[ True False False]]

[[ True False False]
[ True False False]
[ True False False]
...

and the error output:
cv2.error: OpenCV(4.9.0) 👎 error: (-5:Bad argument) in function 'imwrite'

Overload resolution failed:

img data type = bool is not supported
Expected Ptrcv::UMat for argument 'img'
How to save the occlusion images?