SPGNet: Semantic Prediction Guidance for Scene Parsing

TL;DR

This paper introduces SPGNet, a multi-stage encoder-decoder architecture with a Semantic Prediction Guidance module that re-weights features for improved scene parsing accuracy, achieving state-of-the-art results on Cityscapes.

Contribution

The paper proposes a novel SPG module for multi-stage encoder-decoder networks, enhancing feature re-weighting based on pixel-wise semantic guidance for better segmentation.

Findings

SPGNet outperforms single-stage models with similar complexity.

Achieves 81.1% mIoU on Cityscapes test set.

Multi-stage design with SPG improves segmentation accuracy.

Abstract

Multi-scale context module and single-stage encoder-decoder structure are commonly employed for semantic segmentation. The multi-scale context module refers to the operations to aggregate feature responses from a large spatial extent, while the single-stage encoder-decoder structure encodes the high-level semantic information in the encoder path and recovers the boundary information in the decoder path. In contrast, multi-stage encoder-decoder networks have been widely used in human pose estimation and show superior performance than their single-stage counterpart. However, few efforts have been attempted to bring this effective design to semantic segmentation. In this work, we propose a Semantic Prediction Guidance (SPG) module which learns to re-weight the local features through the guidance from pixel-wise semantic prediction. We find that by carefully re-weighting features across stages, a two-stage encoder-decoder network coupled with our proposed SPG module can significantly outperform its one-stage counterpart with similar parameters and computations. Finally, we report experimental results on the semantic segmentation benchmark Cityscapes, in which our SPGNet attains 81.1% on the test set using only 'fine' annotations.