An End-to-End Network for Panoptic Segmentation

TL;DR

This paper introduces an end-to-end neural network for panoptic segmentation that efficiently predicts both instance and stuff segmentation simultaneously, incorporating a spatial ranking module to handle occlusions, and demonstrates promising results on COCO.

Contribution

The paper presents a novel end-to-end network architecture for panoptic segmentation, integrating instance and stuff segmentation with a spatial ranking module for occlusion handling.

Findings

Achieved promising results on the COCO Panoptic benchmark.

The proposed method outperforms traditional two-stage approaches.

Efficient single-network design reduces complexity and improves performance.

Abstract

Panoptic segmentation, which needs to assign a category label to each pixel and segment each object instance simultaneously, is a challenging topic. Traditionally, the existing approaches utilize two independent models without sharing features, which makes the pipeline inefficient to implement. In addition, a heuristic method is usually employed to merge the results. However, the overlapping relationship between object instances is difficult to determine without sufficient context information during the merging process. To address the problems, we propose a novel end-to-end network for panoptic segmentation, which can efficiently and effectively predict both the instance and stuff segmentation in a single network. Moreover, we introduce a novel spatial ranking module to deal with the occlusion problem between the predicted instances. Extensive experiments have been done to validate the performance of our proposed method and promising results have been achieved on the COCO Panoptic benchmark.