Skip to content

Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction

Notifications You must be signed in to change notification settings

daeyun/object-shapes-cvpr18

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

41 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

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 [email protected]: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.

About

Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published