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

We consider the problem of 3D object pose estimation. While much recent workhas focused on the RGB domain, the reliance on accurately annotated imageslimits their generalizability and scalability. On the other hand, the easilyavailable CAD models of objects are rich sources of data, providing a largenumber of synthetically rendered images. In this paper, we solve this keyproblem of existing methods requiring expensive 3D pose annotations byproposing a new method that matches RGB images to CAD models for object poseestimation. Our key innovations compared to existing work include removing theneed for either real-world textures for CAD models or explicit 3D poseannotations for RGB images. We achieve this through a series of objectives thatlearn how to select keypoints and enforce viewpoint and modality invarianceacross RGB images and CAD model renderings. We conduct extensive experiments todemonstrate that the proposed method can reliably estimate object pose in RGBimages, as well as generalize to object instances not seen during training.

Learning Local RGB-to-CAD Correspondences for Object Pose Estimation