Face attribute recognition system and method based on multiple face regions

Abstract

The invention belongs to the field of face recognition, and specifically relates to a face attribute recognition system and method based on multiple face regions, aiming to solve the problem of low recognition rate caused by ignoring detailed information in the face attribute recognition of a single global face image. High problem, the system of the present invention includes a preprocessing unit, a face feature extraction unit, and a prediction unit; the preprocessing unit detects key points of the face in the input image, performs alignment and cropping of the face image, and obtains the first image and its corresponding face key points; the face feature extraction unit extracts global face features, local face features, and fusion face features; the prediction unit extracts the global face features, the local face features, and the fusion The face features are concatenated to obtain the final classification features, and attribute recognition is performed based on a specific attribute classifier. The present invention can extract more effective face attribute features based on multiple areas of the face, and obtain more accurate face attribute estimation.

Description

technical field [0001] The invention belongs to the field of face recognition, and in particular relates to a face attribute recognition system and method based on multiple face regions. Background technique [0002] Usually face attribute analysis often uses a single global face image as input, so it is easy to ignore local and detailed information, and these details are also crucial for analyzing face attributes. Wrinkles and beards also contain rich age information; for expression attributes, when people smile, the corners of their mouths tend to rise. Therefore, both global and local face region information are very important when analyzing face attributes. How to comprehensively utilize global and local face region information for face attribute recognition and improve recognition rate is a problem to be solved in this field. Contents of the invention [0003] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem of l...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem of low recognition rate caused by ignoring the detail information in the face attribute recognition of a single global face image, the first aspect of the present invention proposes a human-based A face attribute recognition system with multiple face regions, the system includes a preprocessing unit, a face feature extraction unit, and a prediction unit;

Images