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

This paper focuses on the challenging task of learning 3D object surfacereconstructions from single RGB images. Existing methods achieve varyingdegrees of success by using different geometric representations. However, theyall have their own drawbacks, and cannot well reconstruct those surfaces ofcomplex topologies. To this end, we propose in this paper a skeleton-bridged,stage-wise learning approach to address the challenge. Our use of skeleton isdue to its nice property of topology preservation, while being of lowercomplexity to learn. To learn skeleton from an input image, we design a deeparchitecture whose decoder is based on a novel design of parallel streamsrespectively for synthesis of curve- and surface-like skeleton points. We usedifferent shape representations of point cloud, volume, and mesh in ourstage-wise learning, in order to take their respective advantages. We alsopropose multi-stage use of the input image to correct prediction errors thatare possibly accumulated in each stage. We conduct intensive experiments toinvestigate the efficacy of our proposed approach. Qualitative and quantitativeresults on representative object categories of both simple and complextopologies demonstrate the superiority of our approach over existing ones. Wewill make our ShapeNet-Skeleton dataset publicly available.

A Skeleton-bridged Deep Learning Approach for Generating Meshes of Complex Topologies from Single RGB Images