DeeperLab: Single-Shot Image Parser

TL;DR

DeeperLab introduces a fully convolutional, single-shot approach for panoptic segmentation that jointly handles semantic and instance segmentation, achieving competitive accuracy and real-time processing speeds.

Contribution

It presents a novel, fully convolutional single-shot model for panoptic segmentation that simplifies the pipeline and improves processing speed compared to prior multi-stage methods.

Findings

Achieves 31.95% PQ on Mapillary Vistas dataset

Operates at near real-time speed of 22.6 fps on GPU

Introduces the region-based Parsing Covering metric

Abstract

We present a single-shot, bottom-up approach for whole image parsing. Whole image parsing, also known as Panoptic Segmentation, generalizes the tasks of semantic segmentation for 'stuff' classes and instance segmentation for 'thing' classes, assigning both semantic and instance labels to every pixel in an image. Recent approaches to whole image parsing typically employ separate standalone modules for the constituent semantic and instance segmentation tasks and require multiple passes of inference. Instead, the proposed DeeperLab image parser performs whole image parsing with a significantly simpler, fully convolutional approach that jointly addresses the semantic and instance segmentation tasks in a single-shot manner, resulting in a streamlined system that better lends itself to fast processing. For quantitative evaluation, we use both the instance-based Panoptic Quality (PQ) metric and the proposed region-based Parsing Covering (PC) metric, which better captures the image parsing quality on 'stuff' classes and larger object instances. We report experimental results on the challenging Mapillary Vistas dataset, in which our single model achieves 31.95% (val) / 31.6% PQ (test) and 55.26% PC (val) with 3 frames per second (fps) on GPU or near real-time speed (22.6 fps on GPU) with reduced accuracy.