Repeated video verification method and device based on frame features

Abstract

The invention relates to a repeated video verification method, device and equipment based on frame features, a computer program product and a computer readable storage medium, which can be applied to various scenes such as cloud technology, artificial intelligence, block chains, intelligent traffic and auxiliary driving. According to the method provided by the embodiment of the invention, the video is verified in a matching manner, and at least one optimal matching frame of each video frame in the original video in each video frame in the to-be-verified video is determined based on the feature similarity of the frame features among the videos; the mode of determining the relatively optimal matching frame in the limited range does not depend on absolute feature similarity, so that the mode does not depend on a limited threshold value, the importance of the absolute feature similarity of the image in repeated video verification is weakened, and the effectiveness and the accuracy of the verification method are improved. In addition, according to the method, the accuracy of repeated video verification is greatly improved by identifying the time sequence characteristics of the optimal video frame.

Description

technical field [0001] The present disclosure relates to the field of repeated video verification, and more specifically, to a repeated video verification method, device, equipment, computer program product and storage medium. Background technique [0002] In the Internet age, multimedia video, as an important data carrier, plays a huge role in information dissemination. Duplicate video detection can not only reduce the spread of redundant information, but also play a role in protecting copyright and preventing the spread of bad videos. Therefore, duplicate video detection is widely used in business scenarios such as low-quality handling identification and copyright detection on various video platforms . However, the information inflation and the typical adversarial relationship existing between the video detector and the video handler pose challenges to the detection of duplicate videos. [0003] by attached Figure 1A , 1B , 1C, on the one hand, the video transfer party...

AI Technical Summary

Problems solved by technology

However, information inflation and the typical adversarial relationship between video detectors and video handlers pose challenges to the detection of duplicate videos.

[0003] by attached Figure 1A , 1B , 1C, on the one hand, the video transfer party edits the video in various ways to prevent it from being detected, and the video detection party needs to constantly modify the detection model and supplement the detection algorithm, for example, by combining the image similarity of the video frame with the In the verification scheme of comparing fixed thresholds to determine whether the video is duplicated, it is easy to have the problem of low verification effectiveness due to the editing of the video transfer party deliberately reducing the image similarity. On the other hand, different videos may also have pictures. The problem of high similarity in composition

Therefore, most of the duplicate video verification schemes in the prior art require very complex feature extraction and feature matching logic

Images