Focal Length and Object Pose Estimation via Render and Compare

• Published in: arXiv.org
• Main Authors: Ponimatkin, Georgy, Labbé, Yann, Russell, Bryan, Aubry, Mathieu, Sivic, Josef
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 11.04.2022
• Subjects: Cameras; Estimation; Pose estimation; Three dimensional models

We introduce FocalPose, a neural render-and-compare method for jointly estimating the camera-object 6D pose and camera focal length given a single RGB input image depicting a known object. The contributions of this work are twofold. First, we derive a focal length update rule that extends an existin...