Online video system and method integrating AV video communication and AI object recognition

Abstract

The present invention provides an online video system and method that integrates AV video communication and AI object recognition. The system includes a terminal body, a rotatable tripod, and a dual camera mechanism; the dual camera mechanism includes a support frame, a dual camera accommodation cavity, and a main camera , from the camera, gyroscope and main control circuit board; the master camera is set forward, the slave camera is set forward and obliquely downward, the master optical axis of the master camera and the slave optical axis of the slave camera have a A certain included angle; the gyroscope is used to collect the setting angle of the main optical axis of the main camera. The advantages are: (1) adopts dual camera design, supports simultaneous AV video calling and AI object recognition, and is convenient to use; (2) supports angle adjustment to adapt to users of different heights and different applicable scenarios; (3) integrated design, simple and beautiful, High reliability. (4) During AI object recognition, the method of correcting images by perspective transformation is used to improve the accuracy of AI object recognition.

Description

technical field [0001] The invention belongs to the technical field of online video education, and in particular relates to an online video system and method integrating AV video communication and AI object recognition. Background technique [0002] As the Internet becomes more and more widely used, the application of online video education is also increasing. In the prior art, when students conduct online video learning, they generally use devices such as IPADs and notebooks for online learning, that is, through the front cameras of the IPADs and notebooks, video communication between students and training teachers is realized. However, this online education method has the following disadvantages: only video communication can be carried out between the students and the training teacher, so that the training teacher can pay attention to the facial expressions of the students, but when the students are taking notes or answering the test paper at the desk, the training teacher...

AI Technical Summary

Problems solved by technology

However, this online education method has the following disadvantages: only video communication can be carried out between the students and the training teacher, so that the training teacher can pay attention to the facial expressions of the students, but when the students are taking notes or answering the test paper at the desk, the training teacher It is impossible to observe the writing information of students, that is, the function of AI object recognition cannot be realized, which leads to the incomplete grasp of the behavior of students by training teachers, which affects the quality of training and teaching

The disadvantage of this solution is obvious. It cannot perform video communication and AI object recognition at the same time, and requires users to switch frequently;

[0005] (2) Design a mirror to reflect the desktop to the camera. This design can only be recognized by AI and does not have the function of video communication; moreover, the angle is single and the scope of use is small;

However, the external camera is not beautiful, and there are reliability problems

Images