Panoptic Segmentation

TL;DR

This paper introduces panoptic segmentation, a unified task combining semantic and instance segmentation, along with a new metric, and analyzes human and machine performance on this task to advance scene understanding.

Contribution

It proposes the novel task of panoptic segmentation, introduces the panoptic quality (PQ) metric, and provides a comprehensive analysis of performance on multiple datasets.

Findings

The PQ metric effectively captures performance across all classes.

Analysis reveals gaps between human and machine performance.

The work aims to unify and advance scene segmentation research.

Abstract

We propose and study a task we name panoptic segmentation (PS). Panoptic segmentation unifies the typically distinct tasks of semantic segmentation (assign a class label to each pixel) and instance segmentation (detect and segment each object instance). The proposed task requires generating a coherent scene segmentation that is rich and complete, an important step toward real-world vision systems. While early work in computer vision addressed related image/scene parsing tasks, these are not currently popular, possibly due to lack of appropriate metrics or associated recognition challenges. To address this, we propose a novel panoptic quality (PQ) metric that captures performance for all classes (stuff and things) in an interpretable and unified manner. Using the proposed metric, we perform a rigorous study of both human and machine performance for PS on three existing datasets, revealing interesting insights about the task. The aim of our work is to revive the interest of the community in a more unified view of image segmentation.