DeepIM: Deep Iterative Matching for 6D Pose Estimation

TL;DR

DeepIM is a deep neural network that iteratively refines 6D object pose estimates by matching rendered images against observed images, significantly improving accuracy over previous methods.

Contribution

The paper introduces DeepIM, a novel neural network that refines 6D object pose estimates through iterative matching, capable of handling unseen objects and outperforming existing techniques.

Findings

DeepIM achieves large improvements over state-of-the-art methods.

The network can match previously unseen objects.

Iterative refinement enhances pose estimation accuracy.

Abstract

Estimating the 6D pose of objects from images is an important problem in various applications such as robot manipulation and virtual reality. While direct regression of images to object poses has limited accuracy, matching rendered images of an object against the observed image can produce accurate results. In this work, we propose a novel deep neural network for 6D pose matching named DeepIM. Given an initial pose estimation, our network is able to iteratively refine the pose by matching the rendered image against the observed image. The network is trained to predict a relative pose transformation using an untangled representation of 3D location and 3D orientation and an iterative training process. Experiments on two commonly used benchmarks for 6D pose estimation demonstrate that DeepIM achieves large improvements over state-of-the-art methods. We furthermore show that DeepIM is able to match previously unseen objects.