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

As a basic component of SE(3)-equivariant deep feature learning, steerableconvolution has recently demonstrated its advantages for 3D semantic analysis.The advantages are, however, brought by expensive computations on dense,volumetric data, which prevent its practical use for efficient processing of 3Ddata that are inherently sparse. In this paper, we propose a novel design ofSparse Steerable Convolution (SS-Conv) to address the shortcoming; SS-Convgreatly accelerates steerable convolution with sparse tensors, while strictlypreserving the property of SE(3)-equivariance. Based on SS-Conv, we propose ageneral pipeline for precise estimation of object poses, wherein a key designis a Feature-Steering module that takes the full advantage ofSE(3)-equivariance and is able to conduct an efficient pose refinement. Toverify our designs, we conduct thorough experiments on three tasks of 3D objectsemantic analysis, including instance-level 6D pose estimation, category-level6D pose and size estimation, and category-level 6D pose tracking. Our proposedpipeline based on SS-Conv outperforms existing methods on almost all themetrics evaluated by the three tasks. Ablation studies also show thesuperiority of our SS-Conv over alternative convolutions in terms of bothaccuracy and efficiency. Our code is released publicly athttps://github.com/Gorilla-Lab-SCUT/SS-Conv.

Sparse Steerable Convolutions: An Efficient Learning of SE(3)-Equivariant Features for Estimation and Tracking of Object Poses in 3D Space