Denoising Diffusion Probabilistic Models on SO(3) for Rotational AlignmentDownload PDF

Mar 02, 2022 (edited Apr 25, 2022)GTRL 2022 PosterReaders: Everyone
  • Keywords: Diffusion Models, DDPM, Denoising Diffusion Probabilistic Models, Manifolds, Rotation, SO(3), Generative Model, Alignment, Probabilistic diffusion models, Data Distribution, Rotation, SO(3), Manifold Diffusion, Roto-translational alignment, Probabilistic Sampling
  • TL;DR: We make DDPMs work for rotations using distributions defined on SO(3), then use them for rotational alignment tasks.
  • Abstract: Probabilistic diffusion models are capable of modeling complex data distributions on high-dimensional Euclidean spaces for a range applications. However, many real world tasks involve more complex structures such as data distributions defined on manifolds which cannot be easily represented by diffusions on $\mathbb{R}^n$. This paper proposes denoising diffusion models for tasks involving 3D rotations leveraging diffusion processes on the Lie group $SO(3)$ in order to generate candidate solutions to rotational alignment tasks. The experimental results show the proposed $SO(3)$ diffusion process outperforms naïve approaches such as Euler angle diffusion in synthetic rotational distribution sampling and in a 3D object alignment task.
  • Poster: png
1 Reply

Loading