Content filter for information handling systems
AssigneeDell Products L.P.
InventorsWalter Kemp, Marcus Daniel Molner, Kai Wang
Life of this patent
Estimating the term…
Abstract
Systems and methods are provided for content filtering by an Information Handling System (IHS). A digital media signal is received by the IHS and the digital media signal is decomposed to isolate two or more different types of content, with one or more streams isolated from the signal for each of the different types of content. The streams of each type of content are input to a content filtering model. The content filtering model outputs at least one of: displayed portions of the digital media signal and audible portions of the digital media signal, to be filtered. The digital medial signal is regenerated from one or more streams of content that have been filtered based on the outputs of the content filtering model.
Claims17
A method for content filtering by an Information Handling System (IHS), the method comprising:
- receiving a digital media signal;
- decomposing the digital media signal to isolate two or more types of content, with one or more streams isolated from the signal for each of the types of content;
- inputting, to a content filtering model comprising a neural network, the one or more streams of each type of content as inputs to the neural network;
- inputting one or more libraries of exemplar unwanted content, wherein the neural network identifies content to be filtered from the one or more streams of each type of content;
- outputting, by the content filtering model, at least one of:
- displayed portions of the digital media signal and audible portions of the digital media signal, to be filtered;
- and
- regenerating the digital media signal from one or more streams of content that have been filtered based on the outputs of the content filtering model.
An Information Handling System (IHS) comprising:
- one or more processors;
- and
- one or more memory devices coupled to the one or more processors, the memory devices storing computer-readable instructions that, upon execution by the one or more processors, cause the first IHS to:
- receive a digital media signal;
- decompose the digital media signal to isolate two or more types of content, with one or more streams isolated from the signal for each of the types of content;
- input, to a content filtering model comprising a neural network, the one or more streams of each type of content as inputs to the neural network;
- input one or more libraries of exemplar unwanted content, wherein the neural network identifies content to be filtered from the one or more streams of each type of content;
- output, by the content filtering model, at least one of:
- displayed portions of the digital media signal and audible portions of the digital media signal, to be filtered;
- and
- regenerate the digital media signal from one or more streams of content that have been filtered based on the outputs of the content filtering model.
A non-transitory computer-readable storage device having instructions stored thereon for content filtering by a first IHS (Information Handling System), wherein execution of the instructions by one or more processors of the first IHS causes the one or more processors to:
- receive a digital media signal;
- decompose the digital media signal to isolate two or more types of content, with one or more streams isolated from the signal for each of the types of content;
- input, to a content filtering model comprising a neural network, the one or more streams of each type of content as inputs to the neural network;
- input one or more libraries of exemplar unwanted content, wherein the neural network identifies content to be filtered from the one or more streams of each type of content;
- output, by the content filtering model, at least one of:
- displayed portions of the digital media signal and audible portions of the digital media signal, to be filtered;
- and
- regenerate the digital media signal from one or more streams of content that have been filtered based on the outputs of the content filtering model.
Dependencies are inferred from the claim text.