Coupled Iterative Refinement for 6D Multi-Object Pose Estimation

TL;DR

This paper introduces a novel end-to-end differentiable framework for 6D multi-object pose estimation that iteratively refines pose and correspondence, achieving state-of-the-art accuracy on standard benchmarks.

Contribution

It presents a new coupled iterative refinement architecture with a differentiable BD-PnP layer for improved 6D object pose estimation.

Findings

Achieves state-of-the-art accuracy on 6D pose benchmarks.

Effectively removes outliers through iterative refinement.

Demonstrates the effectiveness of geometric knowledge integration.

Abstract

We address the task of 6D multi-object pose: given a set of known 3D objects and an RGB or RGB-D input image, we detect and estimate the 6D pose of each object. We propose a new approach to 6D object pose estimation which consists of an end-to-end differentiable architecture that makes use of geometric knowledge. Our approach iteratively refines both pose and correspondence in a tightly coupled manner, allowing us to dynamically remove outliers to improve accuracy. We use a novel differentiable layer to perform pose refinement by solving an optimization problem we refer to as Bidirectional Depth-Augmented Perspective-N-Point (BD-PnP). Our method achieves state-of-the-art accuracy on standard 6D Object Pose benchmarks. Code is available at https://github.com/princeton-vl/Coupled-Iterative-Refinement.