Descriptellation: Deep Learned Constellation Descriptors

• Published in: arXiv.org 2022-09
• Main Authors: Xing, Chunwei, Sun, Xinyu, Cramariuc, Andrei, Gull, Samuel, Chung, Jen Jen, Cadena, Cesar, Siegwart, Roland, Tschopp, Florian
• Format: Article
• Language: English
• Subjects: Background noise; Datasets; Localization; Object recognition; Occlusion; Topology
• ISSN: 2331-8422

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.