Skip to content

[SIGGRAPH Asia 2023] ReShader: View-Dependent Highlights for Single Image View-Synthesis

License

Notifications You must be signed in to change notification settings

avinashpaliwal/ReShader

Repository files navigation

ReShader

ReShader: View-Dependent Highlights for Single Image View-Synthesis
Avinash Paliwal, Brandon Nguyen, Andrii Tsarov, Nima Khademi Kalantari
SIGGRAPH Asia 2023 (TOG)

Paper Project Page Video


demo

Prerequisites

You can setup the anaconda environment using:

conda env create -f environment.yml
conda activate reshader

Download pretrained models. The following script from 3D Moments will download their pretrained models and RGBD-inpainting networks.

./download.sh

Demos

We provided some examples in the examples/ folder. You can render novel views with view-dependent highlights using:

python renderer.py --input_dir examples/camera/ --config configs/render.txt

Training

Training code and dataset to be added.

Citation

@article{paliwal2023reshader,
author = {Paliwal, Avinash and Nguyen, Brandon G. and Tsarov, Andrii and Kalantari, Nima Khademi},
title = {ReShader: View-Dependent Highlights for Single Image View-Synthesis},
year = {2023},
issue_date = {December 2023},
volume = {42},
number = {6},
journal = {ACM Trans. Graph.},
month = {dec},
articleno = {216},
numpages = {9},
}

Acknowledgement

The novel view synthesis part of the code is borrowed from 3D Moments.

About

[SIGGRAPH Asia 2023] ReShader: View-Dependent Highlights for Single Image View-Synthesis

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published