Glass: Geometric Latent Augmentation for Shape Spaces

Sanjeev Muralikrishnan, Siddhartha Chaudhuri, Noam Aigerman, Vladimir G. Kim, Matthew Fisher, Niloy J. Mitra; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 18552-18561


We investigate the problem of training generative models on very sparse collections of 3D models. Particularly, instead of using difficult-to-obtain large sets of 3D models, we demonstrate that geometrically-motivated energy functions can be used to effectively augment and boost only a sparse collection of example (training) models. Technically, we analyze the Hessian of the as-rigid-as-possible (ARAP) energy to adaptively sample from and project to the underlying (local) shape space, and use the augmented dataset to train a variational autoencoder (VAE). We iterate the process, of building latent spaces of VAE and augmenting the associated dataset, to progressively reveal a richer and more expressive generative space for creating geometrically and semantically valid samples. We evaluate our method against a set of strong baselines, provide ablation studies, and demonstrate application towards establishing shape correspondences. GLASS produces multiple interesting and meaningful shape variations even when starting from as few as 3-10 training shapes. Our code is available at https: //

Related Material

[pdf] [arXiv]
@InProceedings{Muralikrishnan_2022_CVPR, author = {Muralikrishnan, Sanjeev and Chaudhuri, Siddhartha and Aigerman, Noam and Kim, Vladimir G. and Fisher, Matthew and Mitra, Niloy J.}, title = {Glass: Geometric Latent Augmentation for Shape Spaces}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2022}, pages = {18552-18561} }