Closed tianweiy closed 3 years ago
Hi, from the paper, it mentioned that you use https://github.com/kujason/ip_basic for depth completion. Which version of the algorithm are you using for KITTI and Waymo? Gaussian, bilateral, multiscale? Thanks.
Hello!
I ended up using the version of IP Basic inside MonoPSR. This uses the follow parameters:
fill_type = 'multiscale' extrapolate = False blur_type = 'bilateral'
thanks
Hi, from the paper, it mentioned that you use https://github.com/kujason/ip_basic for depth completion. Which version of the algorithm are you using for KITTI and Waymo? Gaussian, bilateral, multiscale? Thanks.