We present NeuSE, a novel Neural SE(3)-Equivariant Embedding for objects, and illustrate how it supports object SLAM for consistent spatial understanding with long-term scene changes. NeuSE is a set of latent object embeddings created from partial object observations. It serves as a compact point cloud surrogate for complete object models, encoding full shape information while transforming SE(3)-equivariantly in tandem with the object in the physical world. With NeuSE, relative frame transforms can be directly derived from inferred latent codes. Our proposed SLAM paradigm, using NeuSE for object shape and pose characterization, can operate independently or in conjunction with typical SLAM systems. It directly infers SE(3) camera pose constraints that are compatible with general SLAM pose graph optimization, while also maintaining a lightweight object-centric map that adapts to real-world changes. Our approach is evaluated on synthetic and real-world sequences featuring changed objects and shows improved localization accuracy and change-aware mapping capability, when working either standalone or jointly with a common SLAM pipeline.
@article{arxiv.2303.07308,
title = {NeuSE: Neural SE(3)-Equivariant Embedding for Consistent Spatial Understanding with Objects},
author = {Jiahui Fu and Yilun Du and Kurran Singh and Joshua B. Tenenbaum and John J. Leonard},
journal= {arXiv preprint arXiv:2303.07308},
year = {2023}
}
Comments
15 Pages and 12 figures. Accepted to RSS 2023. Project webpage: https://neuse-slam.github.io/neuse/