In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes. Key to our construction is the introduction of a geometric distortion criterion, defined directly on the decoded shapes, translating the preservation of the metric on the decoding to the formation of linear paths in the underlying latent space. Our rationale lies in the observation that training samples alone are often insufficient to endow generative models with high fidelity, motivating the need for large training datasets. In contrast, metric preservation provides a rigorous way to control the amount of geometric distortion incurring in the construction of the latent space, leading in turn to synthetic samples of higher quality. We further demonstrate, for the first time, the adoption of differentiable intrinsic distances in the backpropagation of a geodesic loss. Our geometric priors are particularly relevant in the presence of scarce training data, where learning any meaningful latent structure can be especially challenging. The effectiveness and potential of our generative model is showcased in applications of style transfer, content generation, and shape completion.

LIMP: Learning Latent Shape Representations with Metric Preservation Priors / Cosmo, L.; Norelli, A.; Halimi, O.; Kimmel, R.; Rodolà, E.. - 12348:(2020), pp. 19-35. (Intervento presentato al convegno 16th European Conference on Computer Vision, ECCV 2020 tenutosi a Virtual Conference) [10.1007/978-3-030-58580-8_2].

LIMP: Learning Latent Shape Representations with Metric Preservation Priors

Cosmo L.;Norelli A.;Rodolà E.
2020

Abstract

In this paper, we advocate the adoption of metric preservation as a powerful prior for learning latent representations of deformable 3D shapes. Key to our construction is the introduction of a geometric distortion criterion, defined directly on the decoded shapes, translating the preservation of the metric on the decoding to the formation of linear paths in the underlying latent space. Our rationale lies in the observation that training samples alone are often insufficient to endow generative models with high fidelity, motivating the need for large training datasets. In contrast, metric preservation provides a rigorous way to control the amount of geometric distortion incurring in the construction of the latent space, leading in turn to synthetic samples of higher quality. We further demonstrate, for the first time, the adoption of differentiable intrinsic distances in the backpropagation of a geodesic loss. Our geometric priors are particularly relevant in the presence of scarce training data, where learning any meaningful latent structure can be especially challenging. The effectiveness and potential of our generative model is showcased in applications of style transfer, content generation, and shape completion.
2020
16th European Conference on Computer Vision, ECCV 2020
Generative model; Learning shapes; Metric distortion
04 Pubblicazione in atti di convegno::04b Atto di convegno in volume
LIMP: Learning Latent Shape Representations with Metric Preservation Priors / Cosmo, L.; Norelli, A.; Halimi, O.; Kimmel, R.; Rodolà, E.. - 12348:(2020), pp. 19-35. (Intervento presentato al convegno 16th European Conference on Computer Vision, ECCV 2020 tenutosi a Virtual Conference) [10.1007/978-3-030-58580-8_2].
File allegati a questo prodotto
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1480125
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 33
  • ???jsp.display-item.citation.isi??? ND
social impact