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/

Zendy Plus

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

Zendy Tools

Zendy Open

The performance of existing single-view 3D reconstruction methods heavilyrelies on large-scale 3D annotations. However, such annotations are tedious andexpensive to collect. Semi-supervised learning serves as an alternative way tomitigate the need for manual labels, but remains unexplored in 3Dreconstruction. Inspired by the recent success of semi-supervised imageclassification tasks, we propose SSP3D, a semi-supervised framework for 3Dreconstruction. In particular, we introduce an attention-guided prototype shapeprior module for guiding realistic object reconstruction. We further introducea discriminator-guided module to incentivize better shape generation, as wellas a regularizer to tolerate noisy training samples. On the ShapeNet benchmark,the proposed approach outperforms previous supervised methods by clear marginsunder various labeling ratios, (i.e., 1%, 5% , 10% and 20%). Moreover, ourapproach also performs well when transferring to real-world Pix3D datasetsunder labeling ratios of 10%. We also demonstrate our method could transfer tonovel categories with few novel supervised data. Experiments on the popularShapeNet dataset show that our method outperforms the zero-shot baseline byover 12% and we also perform rigorous ablations and analysis to validate ourapproach.

Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors