Rich geometric understanding of the world is an important component of many robotic applications such as planning and manipulation. In this paper, we present a modular pipeline for pose and shape estimation of objects from RGB-D images given their category. The core of our method is a generative shape model, which we integrate with a novel initialization network and a differentiable renderer to enable 6D pose and shape estimation from a single or multiple views. We investigate the use of discretized signed distance fields as an efficient shape representation for fast analysis-by-synthesis optimization. Our modular framework enables multi-view optimization and extensibility. We demonstrate the benefits of our approach over state-of-the-art methods in several experiments on both synthetic and real data. We open-source our approach at https://github.com/roym899/sdfest.
@article{arxiv.2207.04880,
title = {SDFEst: Categorical Pose and Shape Estimation of Objects from RGB-D using Signed Distance Fields},
author = {Leonard Bruns and Patric Jensfelt},
journal= {arXiv preprint arXiv:2207.04880},
year = {2022}
}
Comments
Accepted to IEEE Robotics and Automation Letters (and IROS 2022). Project page: https://github.com/roym899/sdfest