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/

Zendy Plus

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

Zendy Tools

Zendy Open

Recently, RGBD-based category-level 6D object pose estimation has achievedpromising improvement in performance, however, the requirement of depthinformation prohibits broader applications. In order to relieve this problem,this paper proposes a novel approach named Object Level Depth reconstructionNetwork (OLD-Net) taking only RGB images as input for category-level 6D objectpose estimation. We propose to directly predict object-level depth from amonocular RGB image by deforming the category-level shape prior intoobject-level depth and the canonical NOCS representation. Two novel modulesnamed Normalized Global Position Hints (NGPH) and Shape-aware Decoupled DepthReconstruction (SDDR) module are introduced to learn high fidelity object-leveldepth and delicate shape representations. At last, the 6D object pose is solvedby aligning the predicted canonical representation with the back-projectedobject-level depth. Extensive experiments on the challenging CAMERA25 andREAL275 datasets indicate that our model, though simple, achievesstate-of-the-art performance.

Object Level Depth Reconstruction for Category Level 6D Object Pose Estimation From Monocular RGB Image