Method and system for matching groups and friends based on situational photos

Abstract

The invention discloses a method and system for matching groups and friends based on situational photos, and the method comprises the following steps: obtaining a user face picture, and carrying out the recognition and detection of the user face picture; matching to obtain a virtual identity of the user; pushing a scene group conforming to the virtual identity to the user according to the virtual identity; recommending friends in the scene group to the user; wherein the generation of the scene group comprises the following steps: a generative adversarial network is utilized to generate a plurality of theme photo libraries, and each theme photo library corresponds to a scene theme; dividing each scene theme into a plurality of specific scene tags; and enabling each scene label to correspond to one scene group. According to the method and the device, the virtual identity is created by utilizing the face of the user, the preset scene group is matched according to the virtual identity, the friend recommendation is acquired in the scene group, the activity of the whole system is based on the virtual identity, and the privacy protection of the user is relatively good.

Description

technical field [0001] The present application relates to the field of social application software, in particular to a method and system for matching groups and friends based on situational photos. Background technique [0002] Faces have natural social attributes, and related face applications and algorithms have sprung up in an endless stream. The development of artificial intelligence technology represented by generative adversarial networks has opened up a new world for the processing of face pictures. Processing, adversarial neural networks have powerful image generation and style memory capabilities. Among them, face recognition applications are also used in the social field. In the social field, how to enable strangers to quickly establish reliable connections is the most fundamental problem of a social software. At present, the commonly used friend recommendation methods are to recommend people you may know according to the number of common friends, or to obtain use...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present application provides a method and system for matching groups and friends based on scene-based photos, which solves the technical problem of privacy leakage in friend recommendation in the prior art, realizes the use of user faces to create virtual identities, and creates virtual identities based on virtual The identity matches the preset situation group. By obtaining friend recommendations in the situation group, the activities of the entire system are based on virtual identities, which is better for user privacy protection.

Images