[CVPR 2020] 3D Photography using Context-aware Layered Depth Inpainting

=====================================

This fork made for Windows !!

Converted by Eran Feit

=====================================

[Paper] [Project Website] [Google Colab]

We propose a method for converting a single RGB-D input image into a 3D photo, i.e., a multi-layer representation for novel view synthesis that contains hallucinated color and depth structures in regions occluded in the original view. We use a Layered Depth Image with explicit pixel connectivity as underlying representation, and present a learning-based inpainting model that iteratively synthesizes new local color-and-depth content into the occluded region in a spatial context-aware manner. The resulting 3D photos can be efficiently rendered with motion parallax using standard graphics engines. We validate the effectiveness of our method on a wide range of challenging everyday scenes and show fewer artifacts when compared with the state-of-the-arts.

3D Photography using Context-aware Layered Depth Inpainting
Meng-Li Shih, Shih-Yang Su, Johannes Kopf, and Jia-Bin Huang
In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.

Prerequisites

• Windows (tested on Windows 10)
• Anaconda
• Python 3.7 (tested on 3.7.4)
• PyTorch 1.9

Please look and follow the Install-instructions.txt file

git clone https://github.com/vt-vl-lab/3d-photo-inpainting.git cd 3d-photo-inpainting

conda create -n 3DP python=3.7 conda activate 3DP

nstall Pytorch in your conda 3DP enviroment https://pytorch.org/

how to find my Cuda version

• nvcc --version

this is for Cuda 11. Please look for the relevant command for your Cuda version

• conda install pytorch torchvision torchaudio cudatoolkit=11.1 -c pytorch -c conda-forge

Python dependencies:

pip install opencv-python==4.2.0.32 pip install vispy==0.6.4 pip install moviepy==1.0.2 pip install transforms3d==0.3.1 pip install networkx==2.3

pip install Cython==3.0a5 pip install git+https://github.com/pattern-inc/cynetworkx.git

pip install scikit-image pip install pyYaml

pip install PyQt5

Please download the model weight using the following command: Run the ready made mode;WG.bat file from the repository home direcrory

• modelWG.bat

Quick start

Please follow the instructions in this section. This should allow to execute our results. For more detailed instructions, please refer to DOCUMENTATION.md.

Execute

1. Put .jpg files (e.g., test.jpg) into the image folder.
   • E.g., image/moon.jpg
2. Run the following command

   python main.py --config argument.yml

   • Note: The 3D photo generation process usually takes about 2-3 minutes depending on the available computing resources.
3. The results are stored in the following directories:
   • Corresponding depth map estimated by MiDaS
     • E.g. depth/moon.npy, depth/moon.png
     • User could edit depth/moon.png manually.
       • Remember to set the following two flags as listed below if user wants to use manually edited depth/moon.png as input for 3D Photo.
         • depth_format: '.png'
         • require_midas: False
   • Inpainted 3D mesh (Optional: User need to switch on the flag save_ply)
     • E.g. mesh/moon.ply
   • Rendered videos with zoom-in motion
     • E.g. video/moon_zoom-in.mp4
   • Rendered videos with swing motion
     • E.g. video/moon_swing.mp4
   • Rendered videos with circle motion
     • E.g. video/moon_circle.mp4
   • Rendered videos with dolly zoom-in effect
     • E.g. video/moon_dolly-zoom-in.mp4
     • Note: We assume that the object of focus is located at the center of the image.
4. (Optional) If you want to change the default configuration. Please read DOCUMENTATION.md and modified argument.yml.

License

This work is licensed under MIT License. See LICENSE for details.

If you find our code/models useful, please consider citing our paper:

@inproceedings{Shih3DP20,
  author = {Shih, Meng-Li and Su, Shih-Yang and Kopf, Johannes and Huang, Jia-Bin},
  title = {3D Photography using Context-aware Layered Depth Inpainting},
  booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2020}
}

Acknowledgments

• We thank Pratul Srinivasan for providing clarification of the method Srinivasan et al. CVPR 2019.
• We thank the author of Zhou et al. 2018, Choi et al. 2019, Mildenhall et al. 2019, Srinivasan et al. 2019, Wiles et al. 2020, Niklaus et al. 2019 for providing their implementations online.
• Our code builds upon EdgeConnect, MiDaS and pytorch-inpainting-with-partial-conv