Discrete Point Flow Networks for Efficient Point Cloud Generation - Archive ouverte HAL Access content directly
Conference Papers Year : 2020

Discrete Point Flow Networks for Efficient Point Cloud Generation

Edmond Boyer
Jakob Verbeek


Generative models have proven effective at modeling 3D shapes and their statistical variations. In this paper we investigate their application to point clouds, a 3D shape representation widely used in computer vision for which, however, only few generative models have yet been proposed. We introduce a latent variable model that builds on normalizing flows with affine coupling layers to generate 3D point clouds of an arbitrary size given a latent shape representation. To evaluate its benefits for shape modeling we apply this model for generation, autoencoding, and single-view shape reconstruction tasks. We improve over recent GAN-based models in terms of most metrics that assess generation and autoencoding. Compared to recent work based on continuous flows, our model offers a significant speedup in both training and inference times for similar or better performance. For single-view shape reconstruction we also obtain results on par with state-of-the-art voxel, point cloud, and mesh-based methods.
Fichier principal
Vignette du fichier
DPF-Nets.pdf (17.87 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02903163 , version 1 (20-07-2020)



Roman Klokov, Edmond Boyer, Jakob Verbeek. Discrete Point Flow Networks for Efficient Point Cloud Generation. ECCV 2020 - 16th European Conference on Computer Vision, Aug 2020, Glasgow, United Kingdom. pp.694-710, ⟨10.1007/978-3-030-58592-1_41⟩. ⟨hal-02903163⟩
266 View
39 Download



Gmail Facebook Twitter LinkedIn More