index.md
---
layout: default
---
# CADSketchNet
>**(2021, Sep 05)** The full paper presentation at the [3DOR2021 - 14th EG 3D Object Retrieval Workshop](https://3dor2021.github.io/index.html) can be viewed [here](https://www.youtube.com/watch?v=snvcuHgJi-k&ab_channel=EG3DOR2021)
>**(2021, Aug 02)** Our paper has been awarded [The Graphics Replicability Stamp](http://www.replicabilitystamp.org/#https-github-com-bharadwaj-manda-cadsketchnet)
>**(2021, July 20)** The final published version of the paper is [here](https://www.sciencedirect.com/science/article/pii/S0097849321001333)
> **(2021, July 03)** The paper is accepted for publication in the Special Section on [3DOR2021 - 14th EG 3D Object Retrieval Workshop](https://3dor2021.github.io/index.html) of the [Computers & Graphics](https://www.journals.elsevier.com/computers-and-graphics) Journal. The arxiv version of the paper is available [here](https://arxiv.org/pdf/2107.06212.pdf).
> **(2021, July 03)** This is the repository for the 'CADSketchNet' Dataset, associated with the paper ['CADSketchNet' - An Annotated Sketch dataset for 3D CAD Model Retrieval with Deep Neural Networks](https://www.sciencedirect.com/science/article/pii/S0097849321001333). For further details, contact Bharadwaj Manda via [here](https://www.linkedin.com/in/bharadwaj-manda-9730ab114/) or [here](bharadwaj-manda.github.io/)
## Dataset
<figure>
<img src="comp-gen.png" alt="data" width="500" height="300"/>
<figcaption>Computer Generated Sketches - Dataset_A</figcaption>
</figure>
<figure>
<img src="hand_drawn.png" alt="data" width="500" height="300"/>
<figcaption>Hand-drawn Sketches - Dataset_B</figcaption>
</figure>
We introduce the CADSketchNet dataset, an annotated collection of sketches of 3D CAD models.
- Dataset-A has 58,696 computer-generated sketches of the 3D CAD models across 68 categories of [MCB](https://mechanical-components.herokuapp.com/)
- Dataset-B has 801 hand-drawn sketches of the 3D CAD models across 42 categories of [ESB](https://engineering.purdue.edu/cdesign/wp/downloads/)
For further details, contact Bharadwaj Manda via [here](https://www.linkedin.com/in/bharadwaj-manda-9730ab114/) or [here](https://bharadwaj-manda.netlify.app/)
## Download
Download the Dataset-A [here](https://drive.google.com/file/d/1kH6Bpt6IWIEOZxcu1vVBK9Bd9WG84kTk/view?usp=sharing)
- Click [here](https://drive.google.com/file/d/1ZUaO4iKDTjz7j9LDSNW8_qpFpqHpDil2/view?usp=sharing) to download the 80-20 train-test split used in the paper
Download the Dataset-B [here](https://drive.google.com/file/d/10XRLO7T58Lq54YtL7wX3CJ1VpsFUo2nW/view?usp=sharing)
### Authors
<a href="bharadwaj-manda.github.io/">Manda, Bharadwaj</a> and <a href="https://www.linkedin.com/in/shubham-dhayarkar-a16a75153/?originalSubdomain=in">Dhayarkar, Shubham</a> and <a href="https://smj007.github.io/">Mitheran, Sai</a> and <a href="https://vkviekash-homepage.github.io/">V.K, Viekash</a> and <a href="https://ed.iitm.ac.in/~raman/">Muthuganapathy, Ramanathan</a>
### Paper/Citation
<a href="https://arxiv.org/pdf/2107.06212.pdf" target="_blank"><img src="paper.png" alt="Paper" class="paper" width="200" height="282"/></a>
This is the arxiv version.
Please cite our paper if you use the CADNET dataset.
```bibtex
@article{MANDA2021100,
title = {‘CADSketchNet’ - An Annotated Sketch dataset for 3D CAD Model Retrieval with Deep Neural Networks},
journal = {Computers & Graphics},
volume = {99},
pages = {100-113},
year = {2021},
issn = {0097-8493},
doi = {https://doi.org/10.1016/j.cag.2021.07.001},
url = {https://www.sciencedirect.com/science/article/pii/S0097849321001333},
author = {Bharadwaj Manda and Shubham Dhayarkar and Sai Mitheran and V.K. Viekash and Ramanathan Muthuganapathy},
keywords = {Retrieval, Search, Dataset, Deep Learning, CAD, Sketch}
}
```
Bharadwaj Manda, Shubham Dhayarkar, Sai Mitheran, V.K. Viekash, Ramanathan Muthuganapathy,
‘CADSketchNet’ - An Annotated Sketch dataset for 3D CAD Model Retrieval with Deep Neural Networks,
Computers & Graphics,
Volume 99, 2021,
Pages 100-113, ISSN 0097-8493,
https://doi.org/10.1016/j.cag.2021.07.001.
(https://www.sciencedirect.com/science/article/pii/S0097849321001333)
### Acknowledgements
We are grateful to the teams of ESB and MCB for making their datasets available. Thanks are also due to the many members who have contributed to the CADSketch dataset.
### License
This dataset is licensed under CC BY-NC-SA: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International This license is one of the Creative Commons licenses and allows users to share the dataset only if they (1) give credit to the copyright holder, (2) do not use the dataset for any commercial purposes, and (3) distribute any additions, transformations or changes to the dataset under this same license.