Pedestrian attribute identification method and identification system based on sequence context relation learning

Abstract

The invention discloses a pedestrian attribute identification method and identification system based on sequence context relation learning. The pedestrian attribute identification method comprises thefollowing steps: establishing and training a pedestrian identification system; encoding the image to be identified into an image sequence in the vertical direction, and initializing an attribute sequence into a random value; calculating an image context relation sequence and an attribute context relation sequence by using the trained pedestrian recognition system; calculating the attention of theimage context relation sequence to each element in the attribute context relation sequence; and calculating the probability that each attribute belongs to each category of the attribute, and selecting the category with the maximum probability value as the category of the attribute. According to the method, the contextual relationship between the image sequences, the contextual relationship between the attributes and the contextual relationship between the images and the attributes are fully utilized, and the accuracy of pedestrian attribute recognition is improved.

Description

technical field [0001] The invention belongs to the technical field of pedestrian attribute recognition, and in particular relates to a pedestrian recognition method and system combining intra-class and inter-class relationships between images and attributes. Background technique [0002] The task of pedestrian attribute recognition is to predict the attribute labels of pedestrians in images, including age, gender, and color of clothes. These attributes contain rich semantic information that can describe the appearance of pedestrians, can bring useful information to pedestrian recognition tasks, and have high application value, so they have gained widespread attention. The main difficulty lies in the change of pedestrian angle and photo illumination, and the long distance will affect the recognition accuracy. [0003] In order to submit the accuracy of pedestrian attribute recognition, most of the current methods input a whole image into the classification network, and make...

AI Technical Summary

Problems solved by technology

Therefore, some methods use manual segmentation of images or manual grouping of attributes to obtain the contextual relationship between elements between images or attributes, but these methods need to use prior knowledge to divide the image into a fixed number of blocks or attributes. Divided into fixed groups, such a fixed grouping makes the learning of the contextual relationship between attributes limited and somewhat unreasonable, and the universality is low. When new attributes are added, they need to be regrouped

Images