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

Category-level object pose estimation involves estimating the 6D pose and the3D metric size of objects from predetermined categories. While recentapproaches take categorical shape prior information as reference to improvepose estimation accuracy, the single-stage network design and training mannerlead to sub-optimal performance since there are two distinct tasks in thepipeline. In this paper, the advantage of two-stage pipeline over single-stagedesign is discussed. To this end, we propose a two-stage deformation-andregistration pipeline called DR-Pose, which consists of completion-aideddeformation stage and scaled registration stage. The first stage uses a pointcloud completion method to generate unseen parts of target object, guidingsubsequent deformation on the shape prior. In the second stage, a novelregistration network is designed to extract pose-sensitive features and predictthe representation of object partial point cloud in canonical space based onthe deformation results from the first stage. DR-Pose produces superior resultsto the state-of-the-art shape prior-based methods on both CAMERA25 and REAL275benchmarks. Codes are available at https://github.com/Zray26/DR-Pose.git.

DR-Pose: A Two-stage Deformation-and-Registration Pipeline for Category-level 6D Object Pose Estimation