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

02/25/2019
by   Chengju Zhou, et al.
0

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 obtain competitive performance on CityPersons dataset while maintaining high computational efficiency.

READ FULL TEXT

page 3

page 5

research
02/27/2018

Fusion of Multispectral Data Through Illumination-aware Deep Neural Networks for Pedestrian Detection

Multispectral pedestrian detection has received extensive attention in r...
research
06/26/2017

Illuminating Pedestrians via Simultaneous Detection & Segmentation

Pedestrian detection is a critical problem in computer vision with signi...
research
05/22/2023

Bright Channel Prior Attention for Multispectral Pedestrian Detection

Multispectral methods have gained considerable attention due to their pr...
research
01/25/2015

Exploring Human Vision Driven Features for Pedestrian Detection

Motivated by the center-surround mechanism in the human visual attention...
research
11/29/2014

Pedestrian Detection aided by Deep Learning Semantic Tasks

Deep learning methods have achieved great success in pedestrian detectio...
research
08/14/2018

Multispectral Pedestrian Detection via Simultaneous Detection and Segmentation

Multispectral pedestrian detection has attracted increasing attention fr...
research
12/02/2022

Dunhuang murals contour generation network based on convolution and self-attention fusion

Dunhuang murals are a collection of Chinese style and national style, fo...

Please sign up or login with your details

Forgot password? Click here to reset