Browse Prior Art Database

System and Method for Advanced Video Filtering Disclosure Number: IPCOM000240860D
Publication Date: 2015-Mar-06
Document File: 3 page(s) / 91K

Publishing Venue

The Prior Art Database


Disclosed is a system that enables video filtering at a frame level based on user sentiment. The system automatically filters a video sequence based on the triangulation of expected video sequence events and user preferences.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

System and Method for Advanced Video Filtering

Viewer access media content (i.e. images, audio, and video) through a number of different devices (e.g., computer, television, mobile device, etc.) from a variety of sources. Many feeds and content include disclaimers regarding the level of propriety for some audiences, such as advising viewer discretion for certain ages. These are "passive" warnings and

do not provide any actual filtering, nor do the warnings account for individual user sentiments. Social media sentiment analysis is also limited; the major focus is around text documents or video transcripts. None of these solutions considers contextual, direct sentiment response from the user to take any proactive filtering action.

A user can express sentiment in the form of a like or dislike of a media content as a whole; however there is no way for a user to express sentiment for a portion of the content. With the only input options being like or dislike, users lack options to express a range of sentiment.

Current art exists that offers media tagging as a whole, whether it be audio, still images or video. However, these fall short on partial, fragmented tagging. Furthermore, there is no contextual association between the actual media content, the tag, and the user's sentiment. Additionally, no method exists to store contextually mapped sentiment data within a database system, and then leverage it for other applications, processes, and services. Any tagging system and storage

repository needs a way to conveniently and accurately store not only content media, but also the users' (or cohort's)

sentiment based on the content that they have consumed.

As media services are used by more consumers, a method is needed that not only allows the end user to articulate sentiment to media content, but also makes a system aware of the...