Skip to content

Code for Ditto: Building Digital Twins of Articulated Objects from Interaction


Notifications You must be signed in to change notification settings


Repository files navigation

Ditto: Building Digital Twins of Articulated Objects from Interaction

Zhenyu Jiang, Cheng-Chun Hsu, Yuke Zhu

CVPR 2022, Oral

Project | arxiv



2022-04-28: We released the data generation code of Ditto here.


Ditto (Digital Twins of Articulated Objects) is a model that reconstructs part-level geometry and articulation model of an articulated object given observations before and after an interaction. Specifically, we use a PointNet++ encoder to encoder the input point cloud observations, and fuse the subsampled point features with a simple attention layer. Then we use two independent decoders to propagate the fused point features into two sets of dense point features, for geometry reconstruction and articulation estimation separately. We construct feature grid/planes by projecting and pooling the point features, and query local features from the constructed feature grid/planes. Conditioning on local features, we use different decoders to predict occupancy, segmentation and joint parameters with respect to the query points. At then end, we can extract explicit geometry and articulation model from the implicit decoders.

If you find our work useful in your research, please consider citing.


  1. Create a conda environment and install required packages.
conda env create -f conda_env_gpu.yaml -n Ditto

You can change the pytorch and cuda version in conda_env_gpu.yaml.

  1. Build ConvONets dependents by running python scripts/ build_ext --inplace.

  2. Download the data, then unzip the under the repo's root.


# single GPU
python experiment=Ditto_s2m

# multiple GPUs
python trainer.gpus=4 +trainer.accelerator='ddp' experiment=Ditto_s2m

# multiple GPUs + wandb logging
python trainer.gpus=4 +trainer.accelerator='ddp' logger=wandb experiment=Ditto_s2m


# only support single GPU
python experiment=Ditto_s2m trainer.resume_from_checkpoint=/path/to/trained/model/


Here is a minimum demo that starts from multiview depth maps before and after interaction and ends with a reconstructed digital twin. To run the demo, you need to install this library for visualization.

We provide the posed depth images of a real word laptop to run the demo. You can download from here and put it under data. You can also run demo your own data that follows the same format.

Data and pre-trained models

Data: here. Remeber to cite Shape2Motion and Abbatematteo et al. as well as Ditto when using these datasets.

Pre-trained models: Shape2Motion dataset, Synthetic dataset.

Useful tips

  1. Run eval "$(python -sc install=bash)" under the root directory, you can have auto-completion for commandline options.

  2. Install pre-commit hooks by pip install pre-commit; pre-commit install, then you can have automatic formatting before each commit.

Related Repositories

  1. Our code is based on this fantastic template Lightning-Hydra-Template.

  2. We use ConvONets as our backbone.


   title={Ditto: Building Digital Twins of Articulated Objects from Interaction},
   author={Jiang, Zhenyu and Hsu, Cheng-Chun and Zhu, Yuke},
   booktitle={arXiv preprint arXiv:2202.08227},