Dual-channel depression angle face fusion correction GAN network and face fusion correction method

Abstract

The invention discloses a dual-channel depression angle face fusion correction GAN network and a face fusion correction method, and the GAN network reconstructs a clear front face through employing the global structure of a low-resolution front face and the local texture of a high-resolution depression angle face, and improves the precision of a face recognition system. The established GAN networkcomprises a super-resolution reconstruction network, an attitude correction network, a head attitude estimation module, a face registration module, a face integration module and other main function modules. The method comprises the following steps: firstly, improving a low-resolution front face to the same resolution as a high-resolution depression angle face through a super-resolution reconstruction network; estimating a head overlooking angle; completing overlooking attitude correction of a high-resolution face through an attitude correction network; realizing pixel-level alignment of the high-resolution face and the high-resolution face by using an optical flow registration method; and finally converting the estimated head overlooking angle into a fusion weight to perform angle-adaptive face synthesis. According to the method, the clear front face can be accurately reconstructed, and a new thought is provided for monitoring video face recognition.

Description

technical field [0001] The invention belongs to the technical field of face recognition, and relates to a face fusion correction network and a face fusion correction method, in particular to a dual-channel depression angle face fusion correction GAN network (generative confrontation network) and a face fusion correction method. [0002] technical background [0003] In public video surveillance situations, since the camera is suspended from a height and shoots downwards, when the target approaches the camera, the face is captured at a top view angle. At this time, the face image has a higher spatial resolution than that of a face at a long-distance front view angle. , but there is often a self-occlusion phenomenon in which half of the face blocks the lower half. In extreme cases, only the image of the forehead can be collected, which brings challenges to normal face recognition. Multi-pose face recognition has always been a research difficulty and hotspot in the field of face...

AI Technical Summary

Problems solved by technology

However, it is impossible to completely cover all overlooking angles with the face samples collected in actual work. If only a limited number of discrete angles are used for training, it is difficult to meet the requirements of arbitrary continuous viewing angle changes during use; improving the completeness of the samples is a huge challenge. cost of

Images