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

LiDAR based place recognition is popular for loop closure detection andre-localization. In recent years, deep learning brings improvements to placerecognition by learnable feature extraction. However, these methods degeneratewhen the robot re-visits previous places with large perspective difference. Toaddress the challenge, we propose DeepRING to learn the roto-translationinvariant representation from LiDAR scan, so that robot visits the same placewith different perspective can have similar representations. There are two keysin DeepRING: the feature is extracted from sinogram, and the feature isaggregated by magnitude spectrum. The two steps keeps the final representationwith both discrimination and roto-translation invariance. Moreover, we statethe place recognition as a one-shot learning problem with each place being aclass, leveraging relation learning to build representation similarity.Substantial experiments are carried out on public datasets, validating theeffectiveness of each proposed component, and showing that DeepRING outperformsthe comparative methods, especially in dataset level generalization.

DeepRING: Learning Roto-translation Invariant Representation for LiDAR based Place Recognition