Weakly-supervised learning of mid-level features for pedestrian attribute recognition and localization

Most existing methods for pedestrian attribute recognition in video surveillance can be formulated as a multi-label image classification methodology, while attribute localization is usually disregarded due to the low image qualities and large variations of camera viewpoints and human poses. In this paper, we propose a weakly-supervised learning based approaching to implementing multi-attribute classification and localization simultaneously, without the need of bounding box annotations of attributes. Firstly, a set of mid-level attribute features are discovered by a multi-scale attribute-aware module receiving the outputs of multiple inception layers in a deep Convolution Neural Network (CNN) e.g., GoogLeNet, where a Flexible Spatial Pyramid Pooling (FSPP) operation is performed to acquire the activation maps of attribute features. Subsequently, attribute labels are predicted through a fully-connected layer which performs the regression between the response magnitudes in activation maps and the image-level attribute annotations. Finally, the locations of pedestrian attributes can be inferred by fusing the multiple activation maps, where the fusion weights are estimated as the correlation strengths between attributes and relevant mid-level features. To validate the proposed approach, extensive experiments are performed on the two currently largest pedestrian attribute datasets, i.e.