This paper presents a novel method for instance segmentation of 3D point clouds. The proposed method is called Gaussian Instance Center Network (GICN), which can approximate the distributions of instance centers scattered in the whole scene as Gaussian center heatmaps. Based on the predicted heatmaps, a small number of center candidates can be easily selected for the subsequent predictions with efficiency, including i) predicting the instance size of each center to decide a range for extracting features, ii) generating bounding boxes for centers, and iii) producing the final instance masks. GICN is a single-stage, anchor-free, and end-to-end architecture that is easy to train and efficient to perform inference. Benefited from the center-dictated mechanism with adaptive instance size selection, our method achieves state-of-the-art performance in the task of 3D instance segmentation on ScanNet and S3DIS datasets.

翻译：本文介绍了一种新型方法,例如3D点云的分解。 提议的方法称为高森实例中心网络(GICN),它可以比较作为高森中心热谱图而分散在整个场景的实验中心的分布分布。 根据预测的热谱图,少数中心候选人可以方便地为随后的预测而挑选,包括(一) 预测每个中心的实例大小,以决定提取特征的范围,(二) 为中心制作捆绑盒,以及(三) 制作最后实例面具。 GICN是一个单级、无锚和端对端结构,易于培训和高效地进行推断。我们的方法得益于以适应性实例大小选择的中央专用机制,在扫描网和S3DIS数据集的3D例分解任务中达到了最先进的性能。