Improved object orientation by doing keypoint correspondences iteratively with BD-PnP
Improved object orientation by doing keypoint correspondences iteratively with BD-PnP
Coupled Iterative Refinement for 6D Multi-Object Pose Estimation
arXiv paper abstract https://arxiv.org/abs/2204.12516v1
arXiv PDF paper https://arxiv.org/pdf/2204.12516v1.pdf
... address the task of 6D multi-object pose: given a set of known 3D objects and an RGB or RGB-D input image, ... detect and estimate the 6D pose of each object.
... propose a new approach ... consists of an end-to-end differentiable architecture that makes use of geometric knowledge.
... approach iteratively refines both pose and correspondence in a tightly coupled manner, allowing us to dynamically remove outliers to improve accuracy.
... use a novel differentiable layer to perform pose refinement by solving an optimization problem ... refer to as Bidirectional Depth-Augmented Perspective-N-Point (BD-PnP).
... method achieves state-of-the-art accuracy on standard 6D Object Pose benchmarks. ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments