Manifold Learning in Quotient Spaces - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

Manifold Learning in Quotient Spaces

Abstract

When learning 3D shapes we are usually interested in their intrinsic geometry rather than in their orientation. To deal with the orientation variations the usual trick consists in augmenting the data to exhibit all possible variability, and thus let the model learn both the geometry as well as the rotations. In this paper we introduce a new autoencoder model for encoding and synthesis of 3D shapes. To get rid of undesirable input variability our model learns a manifold in a quotient space of the input space. Typically, we propose to quotient the space of 3D models by the action of rotations. Thus, our quotient autoencoder allows to directly learn in the space of interest, ignoring side information. This is reflected in better performances on reconstruction and interpolation tasks, as our experiments show that our model outperforms a vanilla autoencoder on the well-known Shapenet dataset. Moreover, our model learns a rotation-invariant representation, leading to interesting results in shapes co-alignment. Finally, we extend our quotient autoencoder to quotient by non-rigid transformations.
Fichier principal
Vignette du fichier
Mehr_Manifold_Learning_in_CVPR_2018_paper.pdf (1.6 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-04051232 , version 1 (29-03-2023)

Identifiers

Cite

Eloi Mehr, André Lieutier, Fernando Sanchez Bermudez, Vincent Guitteny, Nicolas Thome, et al.. Manifold Learning in Quotient Spaces. IEEE Conference on Computer Vision and Pattern Recognition, Jun 2018, Salt Lake City, United States. pp.9165-9174, ⟨10.1109/CVPR.2018.00955⟩. ⟨hal-04051232⟩
14 View
3 Download

Altmetric

Share

Gmail Facebook X LinkedIn More