Rethinking Task and Metrics of Instance Segmentation on 3D Point Clouds

TL;DR

This paper introduces a new approach for 3D point cloud instance segmentation that handles large regions efficiently and proposes novel metrics that evaluate performance independently of input size or category recognition.

Contribution

The paper presents a scalable method with O(Np) complexity and introduces category- and size-independent metrics for more accurate evaluation.

Findings

Achieves state-of-the-art performance on existing benchmarks.

Proposes metrics unaffected by input size or category recognition.

Demonstrates robustness of the new metrics in various scenarios.

Abstract

Instance segmentation on 3D point clouds is one of the most extensively researched areas toward the realization of autonomous cars and robots. Certain existing studies have split input point clouds into small regions such as 1m x 1m; one reason for this is that models in the studies cannot consume a large number of points because of the large space complexity. However, because such small regions occasionally include a very small number of instances belonging to the same class, an evaluation using existing metrics such as mAP is largely affected by the category recognition performance. To address these problems, we propose a new method with space complexity O(Np) such that large regions can be consumed, as well as novel metrics for tasks that are independent of the categories or size of the inputs. Our method learns a mapping from input point clouds to an embedding space, where the embeddings form clusters for each instance and distinguish instances using these clusters during testing. Our method achieves state-of-the-art performance using both existing and the proposed metrics. Moreover, we show that our new metric can evaluate the performance of a task without being affected by any other condition.