Object detection method, device and system, and storage medium

Abstract

Embodiments of the invention provide an object detection method, device and system, and a storage medium. The object detection method comprises the steps of acquiring a to-be-processed image; and inputting the to-be-processed image into a neural network, so as to obtain a characteristic image output by the neural network, wherein the characteristic image comprises a position channel and an attribute channel, the position channel comprises at least one group of candidate position information corresponding to at least one candidate position of at least one prediction object in the to-be-processed image respectively, and the attribute channel comprises at least one group of candidate attribute information corresponding to the at least one candidate position respectively. According to the object detection method and device, the position information and the attribute information of the object are obtained at the same time in a way of sharing the characteristic image, so that the calculationamount is greatly reduced; computing resources and storage resources can be effectively saved, and lightweight terminal deployment schemes can be conveniently realized; and in addition, the processing speed of the neural network for each image can be increased, and the real-time processing of images can be realized.

Description

technical field [0001] The present invention relates to the field of image recognition, and more particularly relates to an object detection method, device, system and storage medium. Background technique [0002] Currently, in some application scenarios, it may be necessary to detect attribute information of a target object. For example, in the field of face recognition, it is sometimes necessary to detect attribute information such as the age of the person to whom the face belongs. In the existing attribute detection methods, the position of the target object is detected from the image first, and then the attribute information of the target object is extracted from the image based on the position of the target object. Existing attribute detection methods need to consume more computing resources, which poses a challenge for deployment on embedded terminals. In addition, the existing attribute detection methods make the entire image processing process take a long time and ...

AI Technical Summary

Problems solved by technology

Existing attribute detection methods need to consume more computing resources, which poses challenges for deployment on embedded terminals

In addition, the existing attribute detection methods make the entire image processing process take a long time and are not real-time enough.

Images