Skip to content
Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction
Branch: master
Clone or download
Latest commit ebd7304 Jun 17, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.idea
cpp
docker
docs update adding all uncommited files. cvpr version. Dec 18, 2017
proto
python update adding all uncommited files. cvpr version. Dec 18, 2017
resources update adding all uncommited files. cvpr version. Dec 18, 2017
third_party training rgb vpo Oct 31, 2017
tools Update README.md Jun 17, 2018
.gitignore
.gitmodules training rgb vpo Oct 31, 2017
CMakeLists.txt
Dockerfile passing all tests with cmake. EGL/OpenGL doesnt work with docker. Apps Jul 26, 2017
README.md

README.md

This repository contains the source code for the following paper:

Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction
Daeyun Shin, Charless Fowlkes, Derek Hoiem
In CVPR, 2018   [pdf]   [supplementary material]   [poster]

Project page

https://www.ics.uci.edu/~daeyuns/pixels-voxels-views/

Installation

This code base is not fully deployable yet (see todo list below). But in the meantime, I can point to where you can find relevant code:

Todo

  • Change hard-coded paths to relative paths.
  • Add an environment.yml file for the Anaconda virtual environment.
  • Provide a download scripts to download the datasets.
  • Upload pre-compiled C++ binaries.
  • Upload sqlite database needed for dataset generation.
  • Upload pre-trained weights for the Pytorch model.

In the meantime, if you need to get any specific part working for research, emailing me is probably the fastest. Contact info can be found here

C++

Clone the repository with the --recurse-submodules option to download third party dependencies.

git clone --recurse-submodules git@github.com:daeyun/scene3d.git

Then run the build scripts in third_party/build_scripts to compile the third party libraries if you don't have these dependencies locally installled.

Finally, run this script to build the binaries:

./tools/build_all.sh

Python

Anaconda environment file coming soon.

You can’t perform that action at this time.