This work is based on our paper DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshes, which appeared at the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2020.