Domain adaptive pedestrian re-identification method based on mutual divergence learning

Abstract

The invention discloses a domain adaptive pedestrian re-identification method based on mutual divergence learning. The method comprises the following steps: preparing a pedestrian data set; pre-training the source domain data set, and extracting feature vectors of pictures from the target domain data set; performing density-based clustering on the images of the target domain data set, and taking the number of the cluster as a pseudo label; adding the outliers into a training sample by using an adversarial strategy; mixing the clustered samples and the outliers, sending the mixture into a network, correcting noise of a pseudo tag by adopting mutual divergence learning, inputting a pedestrian image to be queried into a trained pedestrian re-identification model to obtain a pedestrian feature vector to be identified, performing similarity comparison on the pedestrian feature vector to be identified and attribute features in a candidate library, and obtaining a pedestrian re-identification result. According to the invention, the distribution difference between the source domain and the target domain is reduced, the knowledge of the source domain is effectively utilized, and finally, the framework can learn the characteristics with robustness and discrimination.

Description

technical field [0001] The invention applies deep learning and mutual learning to realize unsupervised domain adaptive pedestrian re-identification, belonging to the field of computer vision. Background technique [0002] Pedestrian re-identification (re-ID) aims to establish identity correspondence between different cameras, judge the images of different cameras, or the technology of whether a specific pedestrian exists in a video sequence, which is usually considered as a sub-problem of image retrieval. It has attracted great attention and made impressive progress in the past decade. Pedestrian re-identification technology is widely used to track a person's trajectory in a large area, and it also has high application value in the fields of robotics, intelligent video surveillance, and automatic photo labeling. [0003] At present, compared with the mature face recognition technology, pedestrian re-identification is still a difficult problem in the field of computer vision...

AI Technical Summary

Problems solved by technology

Although the clustering-based UDA methods have made impressive progress, they still suffer from unavoidable false labels due to the limited transferability of source domain features, the agnosticity of target domain image labels, and the imperfection of clustering results. Noise Remains Barrier to Performance Improvement

Images