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

Robotic systems often require precise scene analysis capabilities, especiallyin unstructured, cluttered situations, as occurring in human-made environments.While current deep-learning based methods yield good estimates of object poses,they often struggle with large amounts of occlusion and do not takeinter-object effects into account. Vision as inverse graphics is a promisingconcept for detailed scene analysis. A key element for this idea is a methodfor inferring scene parameter updates from the rasterized 2D scene. However,the rasterization process is notoriously difficult to invert, both due to theprojection and occlusion process, but also due to secondary effects such aslighting or reflections. We propose to remove the latter from the process bymapping the rasterized image into an abstract feature space learned in aself-supervised way from pixel correspondences. Using only a light-weightinverse rendering module, this allows us to refine 6D object pose estimationsin highly cluttered scenes by optimizing a simple pixel-wise difference in theabstract image representation. We evaluate our approach on the challengingYCB-Video dataset, where it yields large improvements and demonstrates a largebasin of attraction towards the correct object poses.

Refining 6D Object Pose Predictions using Abstract Render-and-Compare