A commodity target detection method and device based on eyeball tracking

Abstract

The invention provides a commodity target detection method and device based on eyeball tracking, and the method comprises the steps of collecting a commodity image in a visual field range of a user; determining attention point position information of the user on the commodity image through an eyeball tracking platform; inputting the commodity image and the attention point position information intoa target learning model, wherein the target learning model is used for determining a target commodity area in the commodity image according to the attention point position information; and obtainingthe commodity information in the target commodity area, wherein the commodity information comprises the category of the commodity, so that the information is tracked according to eyeballs; the attention point position data is acquired, the target learning model is employed to carry out commodity target detection, the problems of slow detection speed and low positioning accuracy in the existing commodity image target detection method are solved, the method has the advantages of fast detection speed and high detection accuracy, and the method can be used for carrying out the automatic target detection on the commodity concerned by the attention point of the user in real time.

Description

technical field [0001] The present invention relates to the technical field of computer vision, in particular to a method and device for detecting commodity objects based on eye tracking. Background technique [0002] Human eye tracking has long been used to study individual visual attention, and the most commonly used eye tracking technology is pupil center corneal reflection (pupil center corneal reflection (PCCR)). The principle of PCCR technology is that, captured by the camera of the physical tracking device, the light source illuminates the pupil to form a highly visible reflected image. These images will be used to determine the reflection of the light source in the cornea and pupil, and finally through the cornea and pupil reflection. The vector angle is calculated with other geometric features to obtain the direction of human gaze. [0003] Traditional target detection methods are usually based on artificial feature extraction, which is not robust to changes in the...

AI Technical Summary

Problems solved by technology

[0004] However, the existing target detection method based on deep learning technology has a slow detection speed, and it is difficult to track the user's attention point in real time, and automatically perform fast product detection based on the user's attention point

Images