Shape part Transfer via semantic latent space factorization
| Autor: | Laurent D. Cohen, Raphael Groscot, Leonidas J. Guibas |
|---|---|
| Přispěvatelé: | CEntre de REcherches en MAthématiques de la DEcision (CEREMADE), Centre National de la Recherche Scientifique (CNRS)-Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Université Paris Dauphine-PSL, Université Paris sciences et lettres (PSL) |
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
autoencoder
050101 languages & linguistics Artificial neural network Computer science business.industry pointcloud 05 social sciences [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] Pattern recognition 02 engineering and technology Space (commercial competition) Object (computer science) Autoencoder latent space Factorization [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] 0202 electrical engineering electronic engineering information engineering [MATH.MATH-AP]Mathematics [math]/Analysis of PDEs [math.AP] 020201 artificial intelligence & image processing 0501 psychology and cognitive sciences Segmentation Artificial intelligence business Encoder Mixing (physics) |
| Zdroj: | Geometric Science of Information Geometric Science of Information (GSI2019) Geometric Science of Information (GSI2019), Aug 2019, Toulouse, France. pp.511-519, ⟨10.1007/978-3-030-26980-7_53⟩ Lecture Notes in Computer Science ISBN: 9783030269791 GSI |
| DOI: | 10.1007/978-3-030-26980-7_53⟩ |
| Popis: | International audience; We present a latent space factorization that controls a generative neural network for shapes in a semantic way. Our method uses the segmentation data present in a shapes collection to explicitly factorize the encoder of a pointcloud autoencoder network, replacing it by several sub-encoders. This allows to learn a semantically-structured latent space in which we can uncover statistical modes corresponding to semantically similar shapes, as well as mixing parts from several objects to create hybrids and quickly exploring design ideas through varying shape combinations. Our work differs from existing methods in two ways: first, it proves the usefulness of neural networks to achieve shape combinations and second, adapts the whole geometry of the object to accommodate for its different parts. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|