Person re identification ( Re ID) refers to retrieving a specific person from a large number of pedestrian images collected by different cameras, which is an important sub topic in the field of computer vision after face recognition. In recent years, great achievements have been made in apply ing deep learning to person re identification. In order to let readers know more about the research pro gress of pedestrian re identification at home and abroad, this paper introduces the relevant literature at home and abroad in recent years and makes analysis and summary. Firstly, person re identification is introduced, and then the basic methods of person re identification based on deep learning are summa rized into three directions: supervised learning, weakly supervised learning, and cross modal learn ing , and their status quo is discussed. Common performance indicators are listed, as well as common ly used datasets and the current best performance on these datasets. After that, the existing problems of person re identification are expounded, and finally, the future research direction has prospected.