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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Constructive Solid Geometry (CSG) is a geometric modeling technique thatdefines complex shapes by recursively applying boolean operations on primitivessuch as spheres and cylinders. We present CSGNe, a deep network architecturethat takes as input a 2D or 3D shape and outputs a CSG program that models it.Parsing shapes into CSG programs is desirable as it yields a compact andinterpretable generative model. However, the task is challenging since thespace of primitives and their combinations can be prohibitively large. CSGNeuses a convolutional encoder and recurrent decoder based on deep networks tomap shapes to modeling instructions in a feed-forward manner and issignificantly faster than bottom-up approaches. We investigate twoarchitectures for this task --- a vanilla encoder (CNN) - decoder (RNN) andanother architecture that augments the encoder with an explicit memory modulebased on the program execution stack. The stack augmentation improves thereconstruction quality of the generated shape and learning efficiency. Ourapproach is also more effective as a shape primitive detector compared to astate-of-the-art object detector. Finally, we demonstrate CSGNet can be trainedon novel datasets without program annotations through policy gradienttechniques.

Neural Shape Parsers for Constructive Solid Geometry