SSA-CNN: Semantic Self-Attention CNN for Pedestrian Detection

TL;DR

This paper introduces SSA-CNN, a multi-task network that leverages semantic segmentation as self-attention cues to enhance pedestrian detection accuracy and efficiency.

Contribution

The paper presents a novel multi-task learning approach that integrates semantic segmentation with pedestrian detection using self-attention mechanisms across multiple scales.

Findings

Achieves 6.27% MR on Caltech dataset, outperforming previous methods.

Effectively suppresses background and highlights pedestrian regions.

Maintains high computational efficiency while improving detection performance.

Abstract

Pedestrian detection plays an important role in many applications such as autonomous driving. We propose a method that explores semantic segmentation results as self-attention cues to significantly improve the pedestrian detection performance. Specifically, a multi-task network is designed to jointly learn semantic segmentation and pedestrian detection from image datasets with weak box-wise annotations. The semantic segmentation feature maps are concatenated with corresponding convolution features maps to provide more discriminative features for pedestrian detection and pedestrian classification. By jointly learning segmentation and detection, our proposed pedestrian self-attention mechanism can effectively identify pedestrian regions and suppress backgrounds. In addition, we propose to incorporate semantic attention information from multi-scale layers into deep convolution neural network to boost pedestrian detection. Experiment results show that the proposed method achieves the best detection performance with MR of 6.27% on Caltech dataset and obtain competitive performance on CityPersons dataset while maintaining high computational efficiency.