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

ing complex 3D shapes with parsimonious part-based representationshas been a long standing goal in computer vision. This paper presents alearning-based solution to this problem which goes beyond the traditional 3Dcuboid representation by exploiting superquadrics as atomic elements. Wedemonstrate that superquadrics lead to more expressive 3D scene parses whilebeing easier to learn than 3D cuboid representations. Moreover, we provide ananalytical solution to the Chamfer loss which avoids the need for computationalexpensive reinforcement learning or iterative prediction. Our model learns toparse 3D objects into consistent superquadric representations withoutsupervision. Results on various ShapeNet categories as well as the SURREALhuman body dataset demonstrate the flexibility of our model in capturing finedetails and complex poses that could not have been modelled using cuboids.

Superquadrics Revisited: Learning 3D Shape Parsing beyond Cuboids