Panoptic-PolarNet: Proposal-free LiDAR Point Cloud Panoptic Segmentation
It addresses the under-explored problem of efficient panoptic segmentation for LiDAR point clouds in urban street scenes, offering a fast and robust solution.
The paper tackles panoptic segmentation for LiDAR point clouds by proposing Panoptic-PolarNet, a framework that unifies semantic segmentation and instance clustering in a single network using a polar BEV representation, achieving 54.1% PQ on SemanticKITTI and leading performance on nuScenes with near real-time speed.
Panoptic segmentation presents a new challenge in exploiting the merits of both detection and segmentation, with the aim of unifying instance segmentation and semantic segmentation in a single framework. However, an efficient solution for panoptic segmentation in the emerging domain of LiDAR point cloud is still an open research problem and is very much under-explored. In this paper, we present a fast and robust LiDAR point cloud panoptic segmentation framework, referred to as Panoptic-PolarNet. We learn both semantic segmentation and class-agnostic instance clustering in a single inference network using a polar Bird's Eye View (BEV) representation, enabling us to circumvent the issue of occlusion among instances in urban street scenes. To improve our network's learnability, we also propose an adapted instance augmentation technique and a novel adversarial point cloud pruning method. Our experiments show that Panoptic-PolarNet outperforms the baseline methods on SemanticKITTI and nuScenes datasets with an almost real-time inference speed. Panoptic-PolarNet achieved 54.1% PQ in the public SemanticKITTI panoptic segmentation leaderboard and leading performance for the validation set of nuScenes.