search

JuewenPeng / BokehMe

BokehMe: When Neural Rendering Meets Classical Rendering (CVPR 2022 Oral)
Apache License 2.0
176 stars 9 forks source link

[question] What method/model do you use for the disparity map? #2

Closed roimulia2 closed 2 years ago

roimulia2 commented 2 years ago

Hey, loved the paper!

I'm curious, what did you use to generate the disparity map? I'm thinking this could be also used to segment the most focused object in a given picture. Could be cool!

JuewenPeng commented 2 years ago

Thank you for loving. We use DPT (https://github.com/isl-org/DPT) to generate demo results due to its good generalization ability. Of course, you can also try other disparity prediction models since our model is not very sensitive to inaccurate disparity inputs.