Multi-feature fusion pedestrian re-identification method based on orientation constraint

The invention relates to a multi-feature fusion pedestrian re-identification method based on orientation constraint, and provides a new network model for orientation change, local occlusion and otherfactors. At present, the research on pedestrian re-identification mostly ignores the influence of orientation difference. According to the invention, while the accuracy is ensured, the target with thesame orientation as a query image can be preferentially identified. According to the method, firstly, a pedestrian orientation classifier is designed, and orientation labeling is carried out on pedestrians in a picture; and then the picture is input into a two-branch convolutional neural network, global features and local features of the pedestrians are extracted for constraint training, one branch processes samples with the same orientation, and the other branch processes samples with different orientations. According to the method, a mixed loss function of orientation constraint is designed, and the network weight i