Content filtering in media playing devices
Various approaches relate to user defined content filtering in media playing devices of undesirable content represented in stored and real-time content from content providers. For example, video, image, and/or audio data can be analyzed to identify and classify content included in the data using various classification models and object and text recognition approaches. Thereafter, the identification and classification can be used to control presentation and/or access to the content and/or portions of the content. For example, based on the classification, portions of the content can be modified (e.g., replaced, removed, degraded, etc.) using one or more techniques (e.g., media replacement, media removal, media degradation, etc.) and then presented.
1. A computing system, comprising:
an input buffer;
an output buffer;
a computing device processor; and
a memory device including instructions that, when executed by the computing device processor, enables the computing system to:
obtain media data from the input buffer,
use a filter that includes a neural network having at least one convolutional layer and at least one temporally recurrent layer to determine a classification of content represented in the media data, wherein the neural network performs an autoencoding to generate a feature vector, sets a value of a feature that represents undesirable content to zero, and performs a decoding of the feature vector,
identify undesirable content based on the classification of the content,
process the media data using the filter to generate filtered media, the undesirable content represented in the filtered media being indiscernible to an observer, and
store the filtered media to the output buffer.
2. The computing system of claim 1 , wherein the media data includes video data, and wherein the instructions, when executed by the computing device processor to determine the classification of the content represented in the media data, further enables the computing system to:
determine a keyframe from the video data using at least one video frame selection algorithm;
analyze the keyframe to identify features representative of the content represented in the keyframe;
determine predetermined features that match the features representative of the content; and
determine the classification of the content based on the predetermined features.
3. The computing system of claim 1 , wherein the filter segments at least one region containing the undesirable content and performs degradation within the at least one region.
4. The computing system of claim 1 , wherein the filter includes a generative neural network, and wherein the generative neural network predicts replacement data at points that, in the media data, provide features that enable an observer to discern the undesirable content, and wherein the filter segments at least one region containing the undesirable content and predicts the replacement data within the at least one region.
5. The computing system of claim 1 , wherein the undesirable content is made indiscernible using a media degradation technique.
6. The computing system of claim 1 , wherein the instructions, when executed by the computing device processor, further enables the computing system to:
obtain restriction preferences associated with a user account; and
classify the content as undesirable based on the restriction preferences.
7. The computing system of claim 1 , wherein the instructions, when executed by the computing device processor, further enables the computing system to:
determine a location and a size of the undesirable content represented in the media data;
determine a replacement object;
scale the replacement object according to the size of the undesirable content; and
overlay the replacement object at the location of the undesirable content.
8. A media playing device, comprising:
an input buffer;
an output buffer;
a computing device processor; and
a memory device including instructions that, when executed by the computing device processor, enables the media playing device to:
obtain media data from the input buffer,
use a filter that includes a neural network having at least one convolutional layer and at least one temporally recurrent layer to determine a classification of content represented in the media data, wherein the neural network performs an autoencoding to generate a feature vector, sets a value of a feature that represents undesirable content to zero, and performs a decoding of the feature vector,
identify undesirable content based on the classification of the content,
process the media data using a filter to generate filtered media, the undesirable content represented in the filtered media being indiscernible to an observer, and
store the filtered media to the output buffer.
9. The media playing device of claim 8 , wherein the filter segments at least one region containing the undesirable content and performs degradation within the at least one region.
10. The media playing device of claim 8 , wherein the filter includes a generative neural network, and wherein the generative neural network predicts replacement data at points that, in the media data, provide features that enable an observer to discern the undesirable content, and wherein the filter segments at least one region containing the undesirable content and predicts the replacement data within the at least one region.
11. The media playing device of claim 8 , wherein the undesirable content is made indiscernible using a media degradation technique.
12. The media playing device of claim 8 , wherein the instructions, when executed by the computing device processor, further enables the media playing device to:
obtain restriction preferences associated with a user account; and
classify the content as undesirable based on the restriction preferences.
13. The media playing device of claim 8 , wherein the instructions, when executed by the computing device processor, further enables the media playing device to:
determine a location and a size of the undesirable content represented in the media data;
determine a replacement object;
scale the replacement object according to the size of the undesirable content; and
overlay the replacement object at the location of the undesirable content.
14. The media playing device of claim 8 , wherein parameters for the filter are stored within the memory device, and wherein the instructions, when executed by the computing device processor, further enables the media playing device to:
obtain updated filter parameters; and
update the filter based on updated filter parameters.
15. The media playing device of claim 8 , wherein the instructions, when executed by the computing device processor, further enables the media playing device to:
identify a set of filter parameters embedded with the media data, wherein the media data is filtered based on the set of filter parameters.
16. The media playing device of claim 8 , wherein the media data includes video data, and wherein the instructions, when executed by the computing device processor to determine the classification of the content represented in the media data, further enables the media playing device to:
determine a keyframe from the video data using at least one video frame selection algorithm;
analyze the keyframe to identify features representative of the content represented in the keyframe;
determine predetermined features that match the features representative of the content; and
determine the classification of the content based on the predetermined features.
17. A non-transitory computer readable storage medium storing instructions that, when executed by at least one processor of a computing system, causes the computing system to:
obtain media data from an input buffer;
use a filter that includes a neural network having at least one convolutional layer and at least one temporally recurrent layer to determine a classification of content represented in the media data, wherein the neural network performs an autoencoding to generate a feature vector, sets a value of a feature that represents undesirable content to zero, and performs a decoding of the feature vector;
identify undesirable content based on the classification of the content;
process the media data using a filter to generate filtered media, the undesirable content represented in the filtered media being indiscernible to an observer; and
store the filtered media to an output buffer.
18. The non-transitory computer readable storage medium of claim 17 , wherein the filter segments at least one region containing the undesirable content and performs degradation within the at least one region.
19. The non-transitory computer readable storage medium of claim 17 , wherein the filter includes a generative neural network, and wherein the generative neural network predicts replacement data at points that, in the media data, provide features that enable an observer to discern the undesirable content, and wherein the filter segments at least one region containing the undesirable content and predicts the replacement data within the at least one region.
20. The non-transitory computer readable storage medium of claim 17 , wherein the undesirable content is made indiscernible using a media degradation technique.
21. The non-transitory computer readable storage medium of claim 17 , wherein the instructions, when executed by the at least one processor, further enables the computing system to:
obtain restriction preferences associated with a user account; and
classify the content as undesirable based on the restriction preferences.
22. The non-transitory computer readable storage medium of claim 17 , wherein the instructions, when executed by the at least one processor, further enables the computing system to:
determine a location and a size of the undesirable content represented in the media data;
determine a replacement object;
scale the replacement object according to the size of the undesirable content; and
overlay the replacement object at the location of the undesirable content.
23. The non-transitory computer readable storage medium of claim 17 , wherein the media data includes video data, and wherein the instructions, when executed by the at least one processor to determine the classification of the content represented in the media data, further enables the computing system to:
determine a keyframe from the video data using at least one video frame selection algorithm;
analyze the keyframe to identify features representative of the content represented in the keyframe;
determine predetermined features that match the features representative of the content; and
determine a classification of the content based on the predetermined features.
24. The non-transitory computer readable storage medium of claim 17 , wherein parameters for the filter are stored within a memory device, and wherein the instructions, when executed by the at least one processor, further enables the computing system to:
obtain updated filter parameters; and
update the filter based on updated filter parameters.
25. The non-transitory computer readable storage medium of claim 17 , wherein the instructions, when executed by the at least one processor, further enables the computing system to:
identify a set of filter parameters embedded with the media data, wherein the media data is filtered based on the set of filter parameters.