Person image recognition method, device, electronic device and computer readable medium

Abstract

Embodiments of the present disclosure disclose a person image recognition method, device, electronic equipment and computer readable medium. A specific implementation of the method includes: acquiring a person image; performing feature extraction on the above-mentioned person image through a feature extraction network included in a preset image recognition model to obtain a global feature map, wherein the above-mentioned preset image recognition model also includes joints A recognition network, wherein the joint recognition network includes a deconvolution branch network group, and the deconvolution branch network group is used to generate a character joint heat map group; the above global feature map is input into the joint recognition network to obtain a character joint heat map group; generating a person image recognition result based on the above-mentioned person joint heat map group, and sending the above-mentioned person image recognition result to a display terminal for display. This implementation manner can improve the accuracy of generating a person image recognition result.

Description

technical field [0001] The embodiments of the present disclosure relate to the field of computer technology, and in particular to a person image recognition method, device, electronic equipment and computer readable medium. Background technique [0002] A character image recognition method is a technology for recognizing character joints in a character image. At present, when recognizing a person image, a common method is to train a model of a general funnel network structure to obtain a model for person image recognition, thereby to recognize the person image and obtain a recognition result. Among them, the model of the general funnel network structure adopts a joint training method in the process of recovering the high-resolution and high-semantic feature map after extracting the low-resolution and high-semantic feature map from the input image, so that each joint corresponds to The feature maps of are extracted by the same network branch and share weights. This can simp...

AI Technical Summary

Problems solved by technology

[0004] First, due to the differences in size and appearance of each joint, and the view fields between different joints are not necessarily correlated, the setting of shared weights is likely to have a negative impact on the final accuracy of each joint, and negative migration occurs. , thus leading to a reduction in the accuracy of generating person image recognition results;

[0005] Second, the predicted heat map generated by the model of the funnel network structure and consistent with the target heat map has absolute information loss due to quantization errors, and it is difficult to determine its exact position on the original image in the subsequent inference process. As a result, the accuracy of generating person image recognition results is reduced

Images