Video scene analysis method and device based on human body

Abstract

The invention discloses a video scene analysis method and a device based on a human body, and the method comprises the steps: collecting an image, carrying out the scene labeling of the image, extracting a human body feature vector of the image through a deep learning method, forming a training sample through the human body feature vector of the image and the labeled scene, and forming a trainingsample set; training a random forest model by utilizing the training set, and obtaining a scene analysis model after parameters of the random forest model are determined; and reading each frame of image of a video to be analyzed, extracting a human body feature vector of the frame of image by using a deep learning method, and calculating and outputting a scene classification result of each frame of image based on the input human body feature vector by using the scene analysis model. According to the method and the device, the scene type of the video can be accurately identified, and the requirements of video automatic analysis and editing on rapid and accurate scene type calculation are met.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a video scene analysis method and device with people as the main body. Background technique [0002] With the continuous enhancement of multimedia technology and the continuous popularization of the Internet, people have put forward higher requirements for the diversity and convenience of capturing information, and the applications and products related to video carriers have increased accordingly, and technologies related to video automation have also emerged as the times require. pregnancy. Technologies such as video automatic analysis, automatic editing, and automatic generation all require accurate calculation and analysis of its attributes and characteristics. [0003] In the process of video creation, various visual elements in various shots will affect the intuitive feeling that the video brings to the audience, and the audience will receive different information and meanin...

AI Technical Summary

Problems solved by technology

This method lacks the recognition of specific semantic features, has a low error tolerance rate, poor robustness, and is not universally applicable to videos of different types and different screen ratios.

[0005] The scene classification of the video screen is related to factors such as the frame, camera angle, human posture, and action angle, so the above-mentioned methods and algorithms are not suitable for fast identification of the video scene. The rate will be very low, which cannot meet the needs of fast and accurate scene calculation and classification in video automation design

Images