Method and device for human face modeling

Abstract

An embodiment of the present invention provides a face modeling method, comprising: obtaining a data source queue through a portrait data distribution module, and forwarding the portrait data in the data source queue to a face modeling queue; The modeling module performs face modeling on the portrait data in the face modeling queue to obtain a modeling result; and reports the storage progress after modeling in real time through the real-time statistics module. The embodiment of the present invention abstracts the warehousing process of face modeling in stages, and performs message communication between stages to achieve the purpose of decoupling. During project implementation, each stage can be reused directly, or slightly modified on top of each stage to meet project requirements. Therefore, this method is universal, can be applied to different projects, avoids modeling for different projects separately, and can reduce labor costs and expenses.

Description

technical field [0001] The present invention relates to the field of face recognition, and more particularly to a method and device for modeling a face. Background technique [0002] Face recognition is a biometric technology for identification based on human facial feature information. Face recognition uses a variety of technologies such as digital images, video processing, pattern recognition, and computer vision. Static ultra-large-scale face recognition retrieval has important applications in the security industry. The general scenario is to perform face recognition retrieval in tens of millions or billions of face databases. Face data of more than tens of millions means that it takes more time to complete face modeling and store in the early stage. However, the existing face retrieval systems often need to customize special modeling and storage programs according to the needs of different projects, and the modeling and storage programs designed for different projects ...

AI Technical Summary

Problems solved by technology

However, the existing face retrieval systems often need to customize special modeling and storage programs according to the needs of different projects, and the modeling and storage programs designed for different projects are not universal

This leads to the need to formulate modeling and warehousing procedures for different projects, which increases labor costs

Images