PointInfinity: Resolution-Invariant Point Diffusion Models

We present PointInfinity, an efficient family of point cloud diffusion models. Our core idea is to use a transformer-based architecture with a fixed-size, resolution-invariant latent representation. This enables efficient training with low-resolution point clouds, while allowing high-resolution poin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 10050 - 10060
Hauptverfasser: Huang, Zixuan, Johnson, Justin, Debnath, Shoubhik, Rehg, James M., Wu, Chao-Yuan
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 16.06.2024
Schlagworte:
3D Diffusion Model
3D Generation
3D reconstruction
3D Vision
Computer vision
Deep Learning
Noise reduction
Point cloud compression
Shape
Three-dimensional displays
Training
Transformer cores
ISSN:1063-6919
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present PointInfinity, an efficient family of point cloud diffusion models. Our core idea is to use a transformer-based architecture with a fixed-size, resolution-invariant latent representation. This enables efficient training with low-resolution point clouds, while allowing high-resolution point clouds to be generated during inference. More importantly, we show that scaling the test-time resolution beyond the training resolution improves the fidelity of generated point clouds and surfaces. We analyze this phenomenon and draw a link to classifier-free guidance commonly used in diffusion models, demonstrating that both allow trading off fidelity and variability during inference. Experiments on CO3D show that PointInfinity can efficiently generate high-resolution point clouds (up to 131k points, 31× more than Point-E) with state-of-the-art quality.
ISSN:1063-6919
DOI:10.1109/CVPR52733.2024.00958