Kubric-NK Dataset

Kubric-NK is a synthetic image dataset based on Kubric designed for evaluating dense prediction models at multiple scales. It was introduced in the CVPR 2025 paper DPFlow: Adaptive Optical Flow Estimation with a Dual-Pyramid Framework to benchmark optical flow estimation models. It provides 30 scenes with 21 frames each at 4 resolutions: 1K, 2K, 4K, and 8K. It also provides the following dense annotations: backward and forward optical flow, depth, surface normals, and object coordinates.

🚀 How to Use the Dataset

The dataset contains multiple .tar files, each containing one set of images or annotations for a particular resolution.

1. Download and Extract

Download the archives for the files and resolution you want (e.g., forward_flow_1k.tar). Extract them into a single root directory.

• You can check the HF Docs about Downloading Datasets.

2. Expected Directory Structure

After extraction, your folder structure will look like this:

forward_flow/
└── 1k/
    ├── 000/
    │   ├── forward_flow_00000.png
    │   ├── forward_flow_00001.png
    │   └── ...
    ├── 001/
    └── ...

📜 Licensing & Acknowledgements

To satisfy both the software and asset requirements of the generation pipeline, this repository operates under the following licenses:

• Dataset (Images & Annotations): The rendered image sequences and annotations in this repository are released under the CC BY 4.0 license.
• Original Kubric Code: The base 3D scenes were generated using a customized fork of Kubric (originally developed by Google Research).
• Kubric-NK Generation Code: The code to generate this dataset is available at https://github.com/hmorimitsu/kubric-nk, and it is released under the Apache 2.0 license.

🎓 Citation

If you use the Kubric-NK dataset in your research, please consider citing the following papers:

@InProceedings{Morimitsu2025DPFlow,
  author    = {Morimitsu, Henrique and Zhu, Xiaobin and Cesar-Jr., Roberto M. and Ji, Xiangyang and Yin, Xu-Cheng},
  booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  title     = {{DPFlow}: Adaptive Optical Flow Estimation with a Dual-Pyramid Framework},
  year      = {2025},
}

@InProceedings{Greff2021kubric,
    title = {Kubric: a scalable dataset generator}, 
    author = {Klaus Greff and Francois Belletti and Lucas Beyer and Carl Doersch and
              Yilun Du and Daniel Duckworth and David J Fleet and Dan Gnanapragasam and
              Florian Golemo and Charles Herrmann and Thomas Kipf and Abhijit Kundu and
              Dmitry Lagun and Issam Laradji and Hsueh-Ti (Derek) Liu and Henning Meyer and
              Yishu Miao and Derek Nowrouzezahrai and Cengiz Oztireli and Etienne Pot and
              Noha Radwan and Daniel Rebain and Sara Sabour and Mehdi S. M. Sajjadi and Matan Sela and
              Vincent Sitzmann and Austin Stone and Deqing Sun and Suhani Vora and Ziyu Wang and
              Tianhao Wu and Kwang Moo Yi and Fangcheng Zhong and Andrea Tagliasacchi},
    booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    year = {2022},
}