Face feature determination method and device, storage medium and electronic device

Abstract

The invention relates to a face feature determination method and device, a storage medium and an electronic device, applied to a GPU, and the method comprises the steps: obtaining a target image in anRGB format; determining resource use information according to the target image, the resource use information comprising a resource reuse relationship for processing the target image and fixed resources required for processing the target image; according to the resource use information, performing face detection and key point detection on the target image, determining face feature information in the target image, and the face feature information comprising a face feature vector and a target position corresponding to the face feature vector. The whole face feature determination process is mainly completed in the GPU, so that switching and interaction between the GPU and the CPU can be reduced, and the operation efficiency is improved.

Description

Technical field [0001] The present disclosure relates to the field of electronic technology, and in particular to a method, device, storage medium and electronic equipment for determining facial features. Background technique [0002] As an important biometric recognition technology, face recognition technology is widely used in finance, national security, justice, education and other fields. The deployment of face recognition systems based on deep learning is generally equipped with higher hardware configuration, especially GPU (English: Graphics Processing Unit, Chinese: Graphics Processor) to ensure that it can meet the real-time requirements in practical applications. In the prior art, the server can realize complete face detection, key point detection, and face feature extraction, but usually the processing sub-modules are independent of each other, and only the network model propagation part is placed on the GPU for execution, so GPUs will be added Interacting with the CPU...

AI Technical Summary

Problems solved by technology

In the existing technology, the server side can realize complete face detection, key point detection and face feature extraction, but usually each processing sub-module is independent of each other, and only the network model propagation part is placed on the GPU for execution, which will increase the number of GPUs. Interact with the CPU (English: Central Processing Unit, Chinese: Central Processing Unit), and at the same time need to switch between the GPU and the CPU, reducing the utilization of the GPU

Images