[drsopp]

Isn't a logical next step to extract the depth field? Possible?

[tetris11]

Isn't the depth decoder part of the processing already?

[drsopp]

I should have read the abstract.

[skywal_l]

Right there in the abstract:

"Our approach works by predicting a dense motion flow field and a monocular depth map directly from a single motion-blurred image"