Integral Ad Science, Inc.

Stefano Vegnaduzzo

Mattia Fumagalli

Kiril Tsemekhman

Anthony J Sabelli

Ana Calabrese

Sergei Izrailev

Mandeep Singh Badwal

Bradley O Timmers

Amy Shriber

Masroor Naved

Igor Benyaminov

Charlie Schumacher

Varun Medappa

Tyler Hughes

Aleks Navratil

Josh Ryan

Methods, systems, and media for generating video classifications using multimodal video analysis are provided. In some embodiments, a method for classifying videos comprising: receiving, from a computing device, a video identifier; parsing a video associated with the video identifier into an audio portion and a plurality of image frames; analyzing the plurality of images frames associated with the video using (i) an optical character recognition technique to obtain first textual information corresponding to text appearing in at least one of the plurality of image frames and (ii) an image classifier to obtain, for each of a plurality of objects appearing in at least one of the plurality of frames of the video, a probability that an object appearing in at least one of the plurality of images falls within an image class; concurrently with analyzing the plurality of image frames associated with the video, analyzing the audio portion of the video using an automated speech recognition technique to obtain second textual information corresponding to words spoken in the video; combining the first textual information, the probability of each of the plurality of objects appearing in the at least one of the plurality of frames of the video, and the second textual information to obtain a combined analysis output for the video; determining, using a neural network, a safety score for each of a plurality of categories that the video contains content belonging to a category of the plurality of categories, wherein the combined analysis output is input into the neural network; and, in response to receiving the video identifier, transmitting a plurality of safety scores corresponding to the plurality of categories to the computing device for the video associated with the video identifier.

Methods, systems, and media for generating video classifications using multimodal video analysis

Methods, systems, and media for generating a media quality score associated with the presentation of a content item are provided. In some embodiments, the method comprises: receiving a plurality of measurement streams associated with a presentation of a video content item comprising a plurality of frames; extracting, using the hardware processor, a plurality of measurements from the plurality of measurement streams; associating each measurement of the plurality of extracted measurements with a particular time position of the presentation of the video content item, wherein groups of measurements are associated with each time position of the presentation of the video content item; determining an interaction score between each measurement in a group of measurements at each particular time position of the plurality of measurement streams to obtain a plurality of interaction scores for the plurality of measurement streams at each time position; combining the plurality of interaction to generate a media quality score for the presentation of the video content item; determining whether the video content item was presented as intended by a content provider; and transmitting a notification that includes the media quality score and that includes an indication of the determination of whether the video content item was presented as intended by the content provider.

Methods, systems, and media for generating a media quality score associated with the presentation of a content item

Growing community of inventors

Ann Arbor, MI, United States of America

Stefano Vegnaduzzo