Skip to content

demo code and pretrained model for ECCV 2018 paper "Reconstruction-based Pairwise Depth Dataset for Depth Image Enhancement Using CNN"

Notifications You must be signed in to change notification settings

JunhoJeon/depth_refine_reconstruct

Repository files navigation

depth_refine_reconstruct

Currently we only provide the code for inference, but the necessary codes for training (network architecture, discontinuity loss) are available now (in ulapsrn.py).

(NEW) clean depth image rendering code updated (2019.05.09)

About

demo code and pretrained model for ECCV 2018 paper "Reconstruction-based Pairwise Depth Dataset for Depth Image Enhancement Using CNN"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published