Panoster: End-To-End Panoptic Segmentation of LiDAR Point Clouds

Gasperini S, Mahani MAN, Marcos-Ramiro A, Navab N, Tombari F (2021)

Publication Type: Journal article

Publication year: 2021

Journal

IEEE Robotics and Automation Letters IEEE

Book Volume: 6

Pages Range: 3216-3223

Article Number: 9357909

Journal Issue: 2

DOI: 10.1109/LRA.2021.3060405

Abstract

Panoptic segmentation has recently unified semantic and instance segmentation, previously addressed separately, thus taking a step further towards creating more comprehensive and efficient perception systems. In this letter, we present Panoster, a novel proposal-free panoptic segmentation method for LiDAR point clouds. Unlike previous approaches relying on several steps to group pixels or points into objects, Panoster proposes a simplified framework incorporating a learning-based clustering solution to identify instances. At inference time, this acts as a class-Agnostic segmentation, allowing Panoster to be fast, while outperforming prior methods in terms of accuracy. Without any post-processing, Panoster reached state-of-The-Art results among published approaches on the challenging SemanticKITTI benchmark, and further increased its lead by exploiting heuristic techniques. Additionally, we showcase how our method can be flexibly and effectively applied on diverse existing semantic architectures to deliver panoptic predictions.

Involved external institutions

Technische Universität München (TUM) DE Germany (DE) BMW AG - Bayerische Motoren Werke / BMW Group DE Germany (DE)

How to cite

APA:

Gasperini, S., Mahani, M.-A.N., Marcos-Ramiro, A., Navab, N., & Tombari, F. (2021). Panoster: End-To-End Panoptic Segmentation of LiDAR Point Clouds. IEEE Robotics and Automation Letters, 6(2), 3216-3223. https://doi.org/10.1109/LRA.2021.3060405

MLA:

Gasperini, Stefano, et al. "Panoster: End-To-End Panoptic Segmentation of LiDAR Point Clouds." IEEE Robotics and Automation Letters 6.2 (2021): 3216-3223.

BibTeX: Download