An Inception-based Variational Autoencoder for Curves Generation and Interpolation
| dc.contributor.author | Barzegar Khalilsaraei, Barzegar Khalilsaraei | |
| dc.contributor.author | Augsdörfer, Ursula | |
| dc.contributor.editor | Skala, Václav | |
| dc.date.accessioned | 2025-07-30T10:17:54Z | |
| dc.date.available | 2025-07-30T10:17:54Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract-translated | In this work, we introduce a novel approach for synthesizing new curves by leveraging Variational Autoencoder (VAE) latent space interpolation. Our method encodes existing ordered point sequ ences representing curves into a compact latent representation, enabling smooth and meaningful transitions between different curve shapes. By performing controlled interpolations in the learned latent space, we generate diverse, high-quality smooth curves that maintain structural coherence and geometric consistency. The proposed method is particularly useful for applications in shape design, procedural modeling, and data augmentation in geometric learning. | en |
| dc.format | 10 s. | cs |
| dc.format.mimetype | application/pdf | |
| dc.identifier.doi | http://www.doi.org/10.24132/CSRN.2025-25 | |
| dc.identifier.issn | 2464-4617 (Print) | |
| dc.identifier.issn | 2464-4625 (online) | |
| dc.identifier.uri | http://hdl.handle.net/11025/62235 | |
| dc.language.iso | en | en |
| dc.publisher | Vaclav Skala - UNION Agency | en |
| dc.rights | © Vaclav Skala - UNION Agency | en |
| dc.rights.access | openAccess | en |
| dc.subject | syntéza tvarů | cs |
| dc.subject | variační autoenkodéry | cs |
| dc.subject | b-spline křivky | cs |
| dc.subject | generativní návrh | cs |
| dc.subject.translated | shape synthesis | en |
| dc.subject.translated | variational autoencoders | en |
| dc.subject.translated | b-spline curves | en |
| dc.subject.translated | generative design | en |
| dc.title | An Inception-based Variational Autoencoder for Curves Generation and Interpolation | en |
| dc.type | konferenční příspěvek | cs |
| dc.type | conferenceObject | en |
| dc.type.status | Peer reviewed | en |
| dc.type.version | publishedVersion | en |
| local.files.count | 1 | * |
| local.files.size | 4618504 | * |
| local.has.files | yes | * |