Perceiving Humans in 3D

MonoLoco++ and MonStereo for 3D localization, orientation, bounding box dimensions and social distancing from monocular and / or stereo images. PyTorch Official Implementation.

  1. Perceiving Humans: from Monocular 3D Localization to Social Distancing (MonoLoco++)
    README & Article



  1. MonStereo: When Monocular and Stereo Meet at the Tail of 3D Human Localization
    README & Article


Both projects has been built upon the CVPR'19 project Openpifpaf
for 2D pose estimation and the ICCV'19 project MonoLoco for monocular 3D localization.
All projects share the AGPL Licence.


Installation steps are the same for both projects.


The installation has been tested on OSX and Linux operating systems, with Python 3.6 or Python 3.7.
Packages have been installed with pip and virtual environments.
For quick installation, do not clone this repository,
and make sure there is no folder named monstereo in your current directory.
A GPU is not required, yet highly recommended for real-time performances.
MonoLoco++ and MonStereo can be installed as a single package, by:

pip3 install monstereo

For development of the monstereo source code itself, you need to clone this repository and then:

pip3 install sdist
cd monstereo
python3 sdist bdist_wheel
pip3 install -e .


All the commands are run through a main file called using subparsers.
To check all the commands for the parser and the subparsers (including openpifpaf ones) run:

  • python3 -m --help
  • python3 -m predict --help
  • python3 -m train --help
  • python3 -m eval --help
  • python3 -m prep --help

or check the file monstereo/

Data structure

├── arrays                 
├── models
├── kitti
├── figures
├── logs

Run the following to create the folders:

mkdir data
cd data
mkdir arrays models kitti figures logs

Further instructions for prediction, preprocessing, training and evaluation can be found here: