Pedestrian Attribute Recognition with Graph Convolutional Network in Surveillance Scenarios

Abstract

Pedestrian attribute recognition is to predict a set of attribute labels of the pedestrian from surveillance scenarios, which is a very challenging task for computer vision due to poor image quality, continual appearance variations, as well as diverse spatial distribution of imbalanced attributes. It is desirable to model the label dependencies between different attributes to improve the recognition performance as each pedestrian normally possesses many attributes. In this paper, we treat pedestrian attribute recognition as multi-label classification and propose a novel model based on the graph convolutional network (GCN). The model is mainly divided into two parts, we first use convolutional neural network (CNN) to extract pedestrian feature, which is a normal operation processing image in deep learning, then we transfer attribute labels to word embedding and construct a correlation matrix between labels to help GCN propagate information between nodes. This paper applies the object classifiers learned by GCN to the image representation extracted by CNN to enable the model to have the ability to be end-to-end trainable. Experiments on pedestrian attribute recognition dataset show that the approach obviously outperforms other existing state-of-the-art methods.

MCGCN: Multi-Correlation Graph Convolutional Network for Pedestrian Attribute Recognition

“…As shown in Table 5, we use four weighting methods on three datasets, where e_w1, e_w2, r_w1, and r_w2 represents the weighting methods in Eqs. ( 24), ( 16), ( 25) and (17), respectively.…”

“…Hence, exploring the semantic correlation among attributes can improve the performance of pedestrian attribute recognition. Several methodologies, which rely on semantic correlation, have been proposed [14]- [17]. Liu et al [14] propose the Attribute Fusion Branch (AFB) module that generates a joint representation for each attribute based on its correlation with other attributes and captures semantic correlations.…”

“…Wang et al [15], [16] explore correlation between different attribute groups through the Recurrent Neural Network (RNN) or Long Short-Term Memory (LSTM) network and predict attributes in a certain order. Song et al [17] implement graph convolution with attribute embedding and prior knowledge of datasets, to learn semantic correlation. However, these methods cannot learn regional correlation.…”

Innovative Topologies and Algorithms for Neural Networks

“…The first paper [1] investigates pedestrian attribute recognition within surveillance scenarios. This challenging task is approached as a form of multi-label classification.…”

Learning a Hierarchical Global Attention for Image Classification

“…By building the network deeper or wider, the accumulation of filters demonstrates a significant improvement in feature representation. Different from fully connection networks, the filters in CNN have fixed receptive fields to concentrate on the local information from features, saving the parameters and making it possible to build the network deeper [3]. Recently, researchers concentrate on well-designed network architectures for efficient feature exploration [4,5].…”