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 work, we present FFB6D, a Full Flow Bidirectional fusion networkdesigned for 6D pose estimation from a single RGBD image. Our key insight isthat appearance information in the RGB image and geometry information from thedepth image are two complementary data sources, and it still remains unknownhow to fully leverage them. Towards this end, we propose FFB6D, which learns tocombine appearance and geometry information for representation learning as wellas output representation selection. Specifically, at the representationlearning stage, we build bidirectional fusion modules in the full flow of thetwo networks, where fusion is applied to each encoding and decoding layer. Inthis way, the two networks can leverage local and global complementaryinformation from the other one to obtain better representations. Moreover, atthe output representation stage, we designed a simple but effective 3Dkeypoints selection algorithm considering the texture and geometry informationof objects, which simplifies keypoint localization for precise pose estimation.Experimental results show that our method outperforms the state-of-the-art bylarge margins on several benchmarks. Code and video are available at\url{https://github.com/ethnhe/FFB6D.git}.

FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation