Descriptellation: Deep Learned Constellation Descriptors

TL;DR

Descriptellation introduces a deep learning-based constellation descriptor for global localization that models semantic object constellations as graphs, demonstrating superior performance and robustness over existing handcrafted descriptors on real-world datasets.

Contribution

It presents a novel deep graph convolution network approach to learn constellation descriptors, improving robustness and generalization in global localization tasks.

Findings

Outperforms state-of-the-art handcrafted descriptors

Shows strong generalization from simulation to real-world data

Robust to environmental noise and perspective changes

Abstract

Current descriptors for global localization often struggle under vast viewpoint or appearance changes. One possible improvement is the addition of topological information on semantic objects. However, handcrafted topological descriptors are hard to tune and not robust to environmental noise, drastic perspective changes, object occlusion or misdetections. To solve this problem, we formulate a learning-based approach by modelling semantically meaningful object constellations as graphs and using Deep Graph Convolution Networks to map a constellation to a descriptor. We demonstrate the effectiveness of our Deep Learned Constellation Descriptor (Descriptellation) on two real-world datasets. Although Descriptellation is trained on randomly generated simulation datasets, it shows good generalization abilities on real-world datasets. Descriptellation also outperforms state-of-the-art and handcrafted constellation descriptors for global localization, and is robust to different types of noise. The code is publicly available at https://github.com/ethz-asl/Descriptellation.