This code accompanies the paper
Layer-structured 3D Scene Inference via View Synthesis
Shubham Tulsiani, Richard Tucker, Noah Snavely
In ECCV, 2018.
Please note that this is not an officially supported Google product.
This is the initial release of the Layered Scene Inference TensorFlow code for learning to infer layered depth images (LDIs) from single input views via multi-view supervision.
You'll first need to follow the installation instructions. To subsequently train and evaluate the LDI prediction models, see the documentation for running experiments using the Synthetic or KITTI datasets.
If you use this code for your research, please consider citing:
@inProceedings{lsiTulsiani18,
title={Layer-structured 3D Scene Inference via View Synthesis},
author = {Shubham Tulsiani
and Richard Tucker
and Noah Snavely},
booktitle={ECCV},
year={2018}
}