English

Neural Isometries: Taming Transformations for Equivariant ML

Computer Vision and Pattern Recognition 2024-10-31 v2 Artificial Intelligence Graphics Machine Learning

Abstract

Real-world geometry and 3D vision tasks are replete with challenging symmetries that defy tractable analytical expression. In this paper, we introduce Neural Isometries, an autoencoder framework which learns to map the observation space to a general-purpose latent space wherein encodings are related by isometries whenever their corresponding observations are geometrically related in world space. Specifically, we regularize the latent space such that maps between encodings preserve a learned inner product and commute with a learned functional operator, in the same manner as rigid-body transformations commute with the Laplacian. This approach forms an effective backbone for self-supervised representation learning, and we demonstrate that a simple off-the-shelf equivariant network operating in the pre-trained latent space can achieve results on par with meticulously-engineered, handcrafted networks designed to handle complex, nonlinear symmetries. Furthermore, isometric maps capture information about the respective transformations in world space, and we show that this allows us to regress camera poses directly from the coefficients of the maps between encodings of adjacent views of a scene.

Keywords

Cite

@article{arxiv.2405.19296,
  title  = {Neural Isometries: Taming Transformations for Equivariant ML},
  author = {Thomas W. Mitchel and Michael Taylor and Vincent Sitzmann},
  journal= {arXiv preprint arXiv:2405.19296},
  year   = {2024}
}

Comments

NeurIPS 2024

R2 v1 2026-06-28T16:45:56.695Z