Browse Prior Art Database

Method of dynamic Realtime Automated Video Tagging of people using the person's Public Social Imprint

IP.com Disclosure Number: IPCOM000254889D
Publication Date: 2018-Aug-10
Document File: 3 page(s) / 157K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Method of dynamic Real-time Automated Video Tagging of people using the person's Public Social Imprint

Background Existing methodologies around the area of dynamic and real-time video tagging does not deal with the real-time approach. Most of the analysis are done based on the historic data and none of them solve the problem of dynamic and Real-time tagging. For e.g: If an individual who is an artist and wants to know the various uploads that have been done that has his video, should be able to get notifications of all the videos that anyone has uploaded on the social media platform. Currently, this capability does not exist in any of the prior art solutions available. Prior Arts US20170318077A1 2017-11-02 METHOD FOR ASSOCIATING PREVIOUSLY CREATED SOCIAL MEDIA DATA WITH AN INDIVIDUAL OR ENTITY (en) This does not cover the checks at real time but is done on historic data. For the trigger points to work, this disclosure requires some information to be updated by the user into the social media platform. Without this manual update from the user, the solution will not be executed. Summary The disclosure brings in a feature that has not been done in any of the existing solutions. This disclosure gives a unique characteristics to the user who takes a video of any other person to automatically tag the individual in the video and alert the potentially tagged individual on the social media platform. The proposed solution is social media platform independent and can be used individually/specifically on the web platforms. This is custom solution which when deployed on to the platform can do the automated monitoring and tagging dynamically. Workflow Steps

1. The solution trigger point is when any user of a social media platform, uploads a

video on the platform.

2. The social media platform by itself, will begin processing the video to make it

compatible with the platform itself. So, the social media platform, while doing the

video processing will split the videos into multiple individual frames. The frames with

clearly visible human faces are identified and images of the same captured for later

processing.

3. Identify the clear images from these marked frames for further processing.

4. The images are then used to search against the social media platform using Face

Recognition/Machine Learning Tools. The API’s will return the matched social

profiles with an accuracy quotient. The tagging of the specific person, will be based

on the threshold crossing of a configurable custom value.

5. Once a defined threshold value is crossed for a particular identified user, he is

tagged privately for the video, on the social media platform.

6. The accuracy of the recognition is a cognitive learning experience based on the user

approval of the identified tags. The user profile identified with the highest accuracy

value above the defined threshhold level, will only be tagge...