English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep generative architectures provide a way to model not only images but alsocomplex, 3-dimensional objects, such as point clouds. In this work, we presenta novel method to obtain meaningful representations of 3D shapes that can beused for challenging tasks including 3D points generation, reconstruction,compression, and clustering. Contrary to existing methods for 3D point cloudgeneration that train separate decoupled models for representation learning andgeneration, our approach is the first end-to-end solution that allows tosimultaneously learn a latent space of representation and generate 3D shape outof it. Moreover, our model is capable of learning meaningful compact binarydescriptors with adversarial training conducted on a latent space. To achievethis goal, we extend a deep Adversarial Autoencoder model (AAE) to accept 3Dinput and create 3D output. Thanks to our end-to-end training regime, theresulting method called 3D Adversarial Autoencoder (3dAAE) obtains eitherbinary or continuous latent space that covers a much wider portion of trainingdata distribution. Finally, our quantitative evaluation shows that 3dAAEprovides state-of-the-art results for 3D points clustering and 3D objectretrieval.

Adversarial Autoencoders for Compact Representations of 3D Point Clouds