diff --git a/README.md b/README.md
index b9647b9..fa6d668 100644
--- a/README.md
+++ b/README.md
@@ -128,6 +128,12 @@ This work introduce ScanObjectNN, a new real-world point cloud object dataset ba
 <p align="center"><img width="50%" src="https://raw.githubusercontent.com/AutodeskAILab/Fusion360GalleryDataset/master/docs/images/reconstruction_teaser.jpg" />
 <img width="50%" src="https://raw.githubusercontent.com/AutodeskAILab/Fusion360GalleryDataset/master/docs/images/segmentation_example.jpg" /></p>
 
+<b>Combinatorial 3D Shape Dataset (2020)</b> [[Link]](https://github.com/POSTECH-CVLab/Combinatorial-3D-Shape-Generation)[[Paper]](https://arxiv.org/abs/2004.07414)
+<br>[Combinatorial 3D Shape Dataset](https://github.com/POSTECH-CVLab/Combinatorial-3D-Shape-Generation) is composed of 406 instances of 14 classes. Each object in our dataset is considered equivalent to a sequence of primitive placement. Compared to other 3D object datasets, our proposed dataset contains an assembling sequence of unit primitives. It implies that we can quickly obtain a sequential generation process that is a human assembling mechanism. Furthermore, we can sample valid random sequences from a given combinatorial shape after validating the sampled sequences. To sum up, the characteristics of our combinatorial 3D shape dataset are (i) combinatorial, (ii) sequential, (iii) decomposable, and (iv) manipulable.
+<p align="center">
+<img width="65%" src="imgs/combinatorial_3d_shape_dataset.png" />
+</p>
+
 <a name="3d_scenes" />
 
 ### 3D Scenes
diff --git a/imgs/combinatorial_3d_shape_dataset.png b/imgs/combinatorial_3d_shape_dataset.png
new file mode 100644
index 0000000..905afd8
Binary files /dev/null and b/imgs/combinatorial_3d_shape_dataset.png differ