Object number recognition model training method and device and storage medium

Abstract

The invention discloses an object number recognition model training method and device and a storage medium. The method comprises the following steps: acquiring a first sample image and a second sample image from a sample image set; inputting the first sample image into the original identification model to obtain a first intermediate feature and a first prediction result, and inputting the second sample image into the original identification model to obtain a second intermediate feature and a second prediction result; adjusting model parameters of the original recognition model according to the first intermediate feature, the second intermediate feature, the first prediction result, the second prediction result, the first true value and the second true value; and under the condition that the recognition accuracy of the original recognition model after model parameter adjustment is higher than a target value, a target recognition model is determined and obtained. The technical problem of poor precision of the number of model identification objects is solved.

Description

technical field [0001] The present invention relates to the field of neural networks, in particular, to a training method, device and storage medium for an object quantity recognition model. Background technique [0002] In the prior art, in the process of recognizing the number of objects in an image, a traditional algorithm is generally used to manually extract image features (eg, edges, textures, gradients, etc.), and then train a regressor from image features to the total number of people. This method lacks high-level semantic information of the image and does not effectively constrain the features, so the recognition accuracy of the trained model is relatively poor. SUMMARY OF THE INVENTION [0003] Embodiments of the present invention provide a training method, device and storage medium for an object quantity recognition model, so as to at least solve the technical problem that the model recognizes the quantity of objects with poor accuracy. [0004] According to an...

AI Technical Summary

Problems solved by technology

[0003] Embodiments of the present invention provide a training method, device, and storage medium for an object number recognition model, so as to at least solve the technical problem of poor accuracy in model recognition of the number of objects

Images