ZeroFOX has patented a system and method for detecting synthetic videos using a combination of weak and strong classifiers. The technology involves training on real and fake videos to identify distinct characteristics and irregularities, enabling accurate detection of fake content. GlobalData’s report on ZeroFOX gives a 360-degree view of the company including its patenting strategy. Buy the report here.

According to GlobalData’s company profile on ZeroFOX, Synthetic data was a key innovation area identified from patents. ZeroFOX's grant share as of April 2024 was 74%. Grant share is based on the ratio of number of grants to total number of patents.

Detecting fake videos using trained classifiers on distinct characteristics

Source: United States Patent and Trademark Office (USPTO). Credit: ZeroFOX Inc

A recently granted patent (Publication Number: US11961282B2) outlines a method and system for training a system to detect fake videos. The method involves creating a training set with real and fake videos, identifying distinct characteristics in each video, extracting these characteristics, down-sampling them, and training a weak classifier to detect irregularities in the characteristics to predict if a test video is fake. The system includes a database with real and fake videos, a server with processors to identify distinct characteristics, extract and down-sample them, and train the weak classifier to detect irregularities and generate a prediction result based on the distinct characteristics.

Furthermore, the method and system involve additional steps such as removing blurred frames and up-sampled frames from fake videos before training the weak classifier, as well as training the weak classifier to detect irregularities in characteristics like head movement, mouth features, and audio gain. The system also includes training multiple weak classifiers to generate prediction results based on different distinct characteristics and down-sampled versions, ultimately leading to the training of a strong classifier to determine the authenticity of test videos based on these prediction results. The strong classifier is trained to assign weights to each prediction result, enhancing the accuracy of detecting fake videos based on the identified distinct characteristics and irregularities.

To know more about GlobalData’s detailed insights on ZeroFOX, buy the report here.

Premium Insights

From

The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.

GlobalData

GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.