Combined learning clothes changing pedestrian re-identification method and system based on identity guidance

• Main Authors: ZHAO YIBO, LI ZHIHUI, GAO ZAN, CHENG ZHIYONG, WEI SHENGXUN, XUE YANBING, ZHUO TAO
• Format: Patent
• Language: Chinese; English
• Published: 16.05.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention belongs to the technical field of computer vision, and provides a joint learning clothes changing pedestrian re-identification method and system based on identity guidance. The method comprises the following steps: acquiring a pedestrian image to be retrieved; inputting the pedestrian image to be retrieved into a pre-trained clothes changing pedestrian re-identification model, and extracting identity robust features irrelevant to clothes as retrieval features; the clothes changing pedestrian re-identification model adaptively weakens interference brought by clothes information through a clothes attention degeneration network guidance model, highlights human body semantic information by using a human body semantic attention and puzzle module, and uses a pedestrian identity enhancement module to guide the model to extract more favorable identity robust representation; and performing similarity matching on the retrieval features and pedestrian image features in a retrieval library. The testing cost