References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: Learnable Earth Parser: Discovering 3D Prototypes in Aerial Scans
(Submitted on 19 Apr 2023 (v1), last revised 28 Mar 2024 (this version, v2))
Abstract: We propose an unsupervised method for parsing large 3D scans of real-world scenes with easily-interpretable shapes. This work aims to provide a practical tool for analyzing 3D scenes in the context of aerial surveying and mapping, without the need for user annotations. Our approach is based on a probabilistic reconstruction model that decomposes an input 3D point cloud into a small set of learned prototypical 3D shapes. The resulting reconstruction is visually interpretable and can be used to perform unsupervised instance and low-shot semantic segmentation of complex scenes. We demonstrate the usefulness of our model on a novel dataset of seven large aerial LiDAR scans from diverse real-world scenarios. Our approach outperforms state-of-the-art unsupervised methods in terms of decomposition accuracy while remaining visually interpretable. Our code and dataset are available at this https URL
Submission history
From: Romain Loiseau [view email][v1] Wed, 19 Apr 2023 14:49:31 GMT (37132kb,D)
[v2] Thu, 28 Mar 2024 17:53:08 GMT (27617kb,D)
Link back to: arXiv, form interface, contact.