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

In this paper, a computation efficient regression framework is presented forestimating the 6D pose of rigid objects from a single RGB-D image, which isapplicable to handling symmetric objects. This framework is designed in asimple architecture that efficiently extracts point-wise features from RGB-Ddata using a fully convolutional network, called XYZNet, and directly regressesthe 6D pose without any post refinement. In the case of symmetric object, oneobject has multiple ground-truth poses, and this one-to-many relationship maylead to estimation ambiguity. In order to solve this ambiguity problem, wedesign a symmetry-invariant pose distance metric, called average (maximum)grouped primitives distance or A(M)GPD. The proposed A(M)GPD loss can make theregression network converge to the correct state, i.e., all minima in theA(M)GPD loss surface are mapped to the correct poses. Extensive experiments onYCB-Video and T-LESS datasets demonstrate the proposed framework'ssubstantially superior performance in top accuracy and low computational cost.

ES6D: A Computation Efficient and Symmetry-Aware 6D Pose Regression Framework