NeuSE: Neural SE(3)-Equivariant Embedding for Consistent Spatial Understanding with Objects
This work addresses the challenge of long-term scene changes in robotics and computer vision, offering an incremental improvement by integrating neural embeddings with existing SLAM pipelines.
The paper tackles the problem of consistent spatial understanding in dynamic scenes by introducing NeuSE, a neural SE(3)-equivariant embedding for objects, which improves localization accuracy and enables change-aware mapping in object SLAM systems.
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.