Learning Category- and Instance-Aware Pixel Embedding for Fast Panoptic Segmentation

TL;DR

This paper introduces a unified, fast panoptic segmentation framework that uses category- and instance-aware pixel embeddings, simplifying the process and achieving competitive results on COCO.

Contribution

It proposes a novel pixel embedding method that encodes semantic and instance information, enabling a one-stage panoptic segmentation approach.

Findings

Achieves fast inference speed.

Performs comparably to two-stage methods on COCO.

First one-stage method with such performance.

Abstract

Panoptic segmentation (PS) is a complex scene understanding task that requires providing high-quality segmentation for both thing objects and stuff regions. Previous methods handle these two classes with semantic and instance segmentation modules separately, following with heuristic fusion or additional modules to resolve the conflicts between the two outputs. This work simplifies this pipeline of PS by consistently modeling the two classes with a novel PS framework, which extends a detection model with an extra module to predict category- and instance-aware pixel embedding (CIAE). CIAE is a novel pixel-wise embedding feature that encodes both semantic-classification and instance-distinction information. At the inference process, PS results are simply derived by assigning each pixel to a detected instance or a stuff class according to the learned embedding. Our method not only demonstrates fast inference speed but also the first one-stage method to achieve comparable performance to two-stage methods on the challenging COCO benchmark.