Object integrity evaluation method, device, electronic device and storage medium

Abstract

An embodiment of the present invention provides a method, device, electronic device, and storage medium for evaluating object integrity. The method includes: inputting an image to be evaluated into a pre-trained convolutional neural network for feature calculation, and extracting the image to be evaluated. Pooling features; inputting the pooling features into a pre-trained first evaluation network, outputting a first evaluation result through the pre-trained first evaluation network; and / or inputting the pooling features into pre-trained The second evaluation network of the second evaluation network outputs a second evaluation result through the pre-trained second evaluation network; based on the first evaluation result and / or the second evaluation result, the object to be evaluated of the image to be evaluated is performed Completeness assessment. The invention can evaluate the completeness of the object to reduce the misrecognition rate of the re-identification model or the attribute recognition model.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an object integrity evaluation method, device, electronic equipment and storage medium. Background technique [0002] In the security field, image recognition based on artificial intelligence occupies an extremely important position, such as re-identification and attribute recognition. Re-identification is a cross-device recognition technology. It can be understood that an image acquisition device collects an object to be evaluated, and by searching the image corresponding to the object to be evaluated in other devices, it can automatically obtain where the object to be evaluated appears. Attribute recognition is a recognition technology based on target detection, such as identifying the color or style of clothes worn by the target person, the tire shape of the target vehicle, etc. However, in real scenes, due to factors such as the installation position ...

AI Technical Summary

Problems solved by technology

However, in real scenes, due to factors such as the installation position of the camera and the pose of the object to be evaluated, the acquisition of the object of interest is incomplete. For example, in the image collected by the image acquisition device, the object to be evaluated is truncated by the edge of the image , or a part of the object to be evaluated is blocked by other objects, and the re-identification model or attribute recognition model recognizes without knowing whether the object to be recognized is complete, which may lead to misidentification

Images