Robust Bayesian Scene Reconstruction with Retrieval-Augmented Priors for Precise Grasping and Planning

TL;DR

This paper introduces BRRP, a probabilistic scene reconstruction method that uses retrieval-augmented priors from mesh datasets to accurately infer 3D object geometries from single RGBD images, improving robustness and manipulation success.

Contribution

We propose a retrieval-augmented prior for Bayesian scene reconstruction that leverages existing mesh datasets to infer occluded object geometries from noisy observations.

Findings

BRRP outperforms deep learning methods in robustness to noise.

BRRP achieves higher accuracy than non-informative prior methods.

Real-world experiments show improved manipulation in cluttered environments.

Abstract

Constructing 3D representations of object geometry is critical for many robotics tasks, particularly manipulation problems. These representations must be built from potentially noisy partial observations. In this work, we focus on the problem of reconstructing a multi-object scene from a single RGBD image using a fixed camera. Traditional scene representation methods generally cannot infer the geometry of unobserved regions of the objects in the image. Attempts have been made to leverage deep learning to train on a dataset of known objects and representations, and then generalize to new observations. However, this can be brittle to noisy real-world observations and objects not contained in the dataset, and do not provide well-calibrated reconstruction confidences. We propose BRRP, a reconstruction method that leverages preexisting mesh datasets to build an informative prior during robust probabilistic reconstruction. We introduce the concept of a retrieval-augmented prior, where we retrieve relevant components of our prior distribution from a database of objects during inference. The resulting prior enables estimation of the geometry of occluded portions of the in-scene objects. Our method produces a distribution over object shape that can be used for reconstruction and measuring uncertainty. We evaluate our method in both simulated scenes and in the real world. We demonstrate the robustness of our method against deep learning-only approaches while being more accurate than a method without an informative prior. Through real-world experiments, we particularly highlight the capability of BRRP to enable successful dexterous manipulation in clutter.