Quantum mechanics

Robot

Algorithm

Scale (ratio)

Linguistics

Engineering

Philosophy

Net (polyhedron)

Feature learning

Simultaneous localization and mapping

Computer science

Physics

Performance improvement

Point cloud

Pattern recognition (psychology)

Mathematics

Politics

Political science

Feature (linguistics)

Mobile robot

Operations management

Robotics

Artificial intelligence

Geometry

Representation (politics)

Point (geometry)

End-to-end principle

Deep learning

Artificial neural network

Consistency (knowledge bases)

Machine learning

State (computer science)

Retrieval-based place recognition is an efficient and effective solution for re-localization within a pre-built map, or global data association for Simultaneous Localization and Mapping (SLAM). The accuracy of such an approach is heavily dependent on the quality of the extracted scene-level representation. While end-to-end solutions - which learn a global descriptor from input point clouds - have demonstrated promising results, such approaches are limited in their ability to enforce desirable properties at the local feature level. In this paper, we introduce a local consistency loss to guide the network towards learning local features which are consistent across revisits, hence leading to more repeatable global descriptors resulting in an overall improvement in 3D place recognition performance. We formulate our approach in an end-to-end trainable architecture called LoGG3D-Net. Experiments on two large-scale public benchmarks (KITTI and MulRan) show that our method achieves mean $F1_{max}$ scores of $0.939$ and $0.968$ on KITTI and MulRan respectively, achieving state-of-the-art performance while operating in near real-time. The open-source implementation is available at: https://github.com/csiro-robotics/LoGG3D-Net.

LoGG3D-Net: Locally Guided Global Descriptor Learning for 3D Place Recognition