Skip to content
Visualize 3DPeople Dataset
Jupyter Notebook
Branch: master
Clone or download
Latest commit 20bd051 Jan 17, 2020
Type Name Latest commit message Commit time
Failed to load latest commit information. Update Jan 16, 2020
vis_dataset.ipynb added jupyter Dec 18, 2019


[Project] [Paper] [Dataset]

First dataset of dressed humans with specific geometry representation for the clothes. It contains ~2 Million images with 40 male/40 female performing 70 actions. Every subject-action sequence is captured from 4 camera views and annotated with: RGB, 3D skeleton, body part and cloth segmentation masks, depth map, optical flow, and camera parameters.

We can NOT share the 3D meshes nor raw files of the models for copyright reasons. The actions used are from the CMU MoCap database.


Train/Test Splits:

  • Train: woman01-woman33 man01-man33
  • Test: woman34-woman40 man34-man40

Visualize data:

Download jupyter notebook. You can use the sample dataset. Full dataset can be download in the dataset website.

conda create -n 3dpeople
source activate 3dpeople
conda install matplotlib opencv pillow scipy
conda install -c conda-forge ipywidgets=7.2.1
conda install -c plotly chart-studio
jupyter nbextension enable --py widgetsnbextension
jupyter notebook vis_dataset.ipynb 

Inside notebook select the sequence to visualize: "seq_name=..." in first cell


If you use this data cite:

    title={{3DPeople: Modeling the Geometry of Dressed Humans}},
    author={Pumarola, Albert and Sanchez, Jordi and Choi, Gary and Sanfeliu, Alberto and Moreno-Noguer, Francesc},
    booktitle={International Conference in Computer Vision (ICCV)},
You can’t perform that action at this time.