Method and device for determining attribute information of target object

Abstract

The invention discloses a method and device for determining attribute information of a target object. A specific embodiment way of the method comprises the steps that video monitoring images of a target area are obtained; the target object in the video monitoring images is detected and the target object comprises people in the target area; feature information of the target object, article information relevant to the target object and scene information relevant to the target object based on the video monitoring images are extracted; analysis and process are conducted on the feature information of the target object, the article information relevant to the target object and the scene information relevant to the target object to determine the attribute information of the target object in the target area. The method and device for determining the attribute information of the target object use relevant information of richer target objects and achieve more precise analysis of the attribute information of target objects.

Description

technical field [0001] The present application relates to the field of computer technology, specifically to the technical field of information retrieval in video surveillance images, and in particular to a method and device for determining attribute information of a target object. Background technique [0002] A user profile can be a collection of user attribute information, and can be a model used to describe user attribute characteristics. At present, user portraits are generally constructed based on behavior log information obtained from terminals associated with the user (such as the user's mobile phone, personal computer, etc.). This type of behavior log information can be customized by the user. For example, the user can enter information such as gender, age, and hobbies on the terminal; it can also be obtained by marking and counting the user's behavior by the terminal, such as the user's network information that can be saved by the terminal. Access records, location...

AI Technical Summary

Problems solved by technology

However, the terminal cannot obtain some actual behavior information of the user, such as the information of using cash to purchase a certain brand of clothing

When the user does not use the location service application, the terminal cannot obtain the user's location information and track information

Therefore, the user information used in the construction of user portraits in the prior art is not comprehensive, and the accuracy of the constructed user portraits needs to be improved

In addition, since the construction of user portraits based on terminals is generally applicable to online applications, in offline application scenarios, it is impossible to analyze the user's offline behavior information in a timely manner, so that it is impossible to make accurate decisions based on the user's real-time offline behavior

Images