English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Denoising diffusion models (DDMs) have shown promising results in 3D pointcloud synthesis. To advance 3D DDMs and make them useful for digital artists,we require (i) high generation quality, (ii) flexibility for manipulation andapplications such as conditional synthesis and shape interpolation, and (iii)the ability to output smooth surfaces or meshes. To this end, we introduce thehierarchical Latent Point Diffusion Model (LION) for 3D shape generation. LIONis set up as a variational autoencoder (VAE) with a hierarchical latent spacethat combines a global shape latent representation with a point-structuredlatent space. For generation, we train two hierarchical DDMs in these latentspaces. The hierarchical VAE approach boosts performance compared to DDMs thatoperate on point clouds directly, while the point-structured latents are stillideally suited for DDM-based modeling. Experimentally, LION achievesstate-of-the-art generation performance on multiple ShapeNet benchmarks.Furthermore, our VAE framework allows us to easily use LION for differentrelevant tasks: LION excels at multimodal shape denoising and voxel-conditionedsynthesis, and it can be adapted for text- and image-driven 3D generation. Wealso demonstrate shape autoencoding and latent shape interpolation, and weaugment LION with modern surface reconstruction techniques to generate smooth3D meshes. We hope that LION provides a powerful tool for artists working with3D shapes due to its high-quality generation, flexibility, and surfacereconstruction. Project page and code: https://nv-tlabs.github.io/LION.

LION: Latent Point Diffusion Models for 3D Shape Generation