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

Planar grasp detection is one of the most fundamental tasks to roboticmanipulation, and the recent progress of consumer-grade RGB-D sensors enablesdelivering more comprehensive features from both the texture and shapemodalities. However, depth maps are generally of a relatively lower qualitywith much stronger noise compared to RGB images, making it challenging toacquire grasp depth and fuse multi-modal clues. To address the two issues, thispaper proposes a novel learning based approach to RGB-D grasp detection, namelyDepth Guided Cross-modal Attention Network (DGCAN). To better leverage thegeometry information recorded in the depth channel, a complete 6-dimensionalrectangle representation is adopted with the grasp depth dedicatedly consideredin addition to those defined in the common 5-dimensional one. The prediction ofthe extra grasp depth substantially strengthens feature learning, therebyleading to more accurate results. Moreover, to reduce the negative impactcaused by the discrepancy of data quality in two modalities, a LocalCross-modal Attention (LCA) module is designed, where the depth features arerefined according to cross-modal relations and concatenated to the RGB ones formore sufficient fusion. Extensive simulation and physical evaluations areconducted and the experimental results highlight the superiority of theproposed approach.

RGB-D Grasp Detection via Depth Guided Learning with Cross-modal Attention