Indian Science Technology and Engineering facilities Map
 
Preivious Next  

  Digital Catalogue for Technology and Products Development


   Technology and Product Development

    Basic Information

Technology developed: TV News Broadcast Video Segmentation System (TP19765102568)
Category: Technology Service/Know how
Details of Inventor(s):
Inventor Institution/Organization/Company Department Designation
RAGHVENDRA KANNAO, PRITHWIJIT GUHA Indian Institute of Technology Guwahati ELECTRONICS AND ELECTRICAL ENGINEERING ASSISTANT PROFESSOR
Technical Application Area: Artificial Intelligence & Machine Learning
If 'Other', please specify:
Please give more details of new technical application area:
COMPUTER VISION MACHINE LEARNING DEEP LEARNING SPEECH and AUDIO SIGNAL PROCESSING PARALLEL PROGRAMMING DATABASES SOFTWARE ARCHITECTURE
Organization(s):
Indian Institute of Technology (IIT) Guwahati
Affiliated Ministry: Ministry of Education
Type of technology development: Indigenous
Does the technology help in replacing any import items currently
procured from outside India?
Yes
Does the technology have export potential? Yes
Category of Technology developed: Futuristic
Stage of Development: Prototype Level
Please describe in detail including the TRL Level:
The proposed TV news broadcast video segmentation system is deployed using two workstations and eleven high end desktops with a NAS for real time segmentation of video data acquired from three different news channels The system is benchmarked on 360 hours of ground truth marked data The status of this work is at TRL 5

    Abstract:

Applications: Segmentation of TV broadcast into semantically meaningful units is a necessary first step for its analysis This work presents novel methods for TV news broadcast segmentation at three semantic levels Segregation of channel content and advertisements Identification of Debates Interviews and News Bulletins in Channel Content Detecting Story Boundaries in News Bulletins This work advocates the usage of presentation styles for the segmentation task To this end novel features derived from shot categories overlay text distribution gradient and audioclass histogram are proposed Three novel feature fusion and classification schemes Progressivelybalanced PerceptronTree Successweighted MultiKernel Learning and SVMEnsemble are presented in this work These generic classifiers are used for various intermediate binary and multiclass classification problems involved in the segmentation task This work also contributes a detailed software architecture for implementing an end to end broadcast segmentation system A 360 hours of news broadcast video dataset is also contributed for benchmarking various segmentation tasks The derivatives of this work has the following applications. Commercial detection in TV broadcast videos News Debate and Interview detection Story Segmentation in News Bulletins Early Intermediate and Late Feature Fusion for General Classification Problems Remotely controllable video recording tool with scheduling Tool for marking and visualizing video segments Tool for labelling images and video shots
Advantages: Commercially available broadcast monitoring products are mostly limited by their dependences on metadata and protocols provided by broadcasting agencies Thus stakeholders have to often resort to manual monitoring and segmentation Considering these market opportunities two commercially viable products can be carved out from this work First an end to end automatic system for broadcast segmentation can be productized The proposed methods are independent of any channel or broadcaster specific assumptions and require minimal manual intervention Thus such an automatic system can be easily scaled up for a large number of TV news channels Moreover the system could be deployed on embedded devices to create standalone broadcast monitoring stations for locally operated news channels The advertisement detection module can also be commercialized and marketed as an independent product Such advertisement detector can be used for deriving competitive business intelligence and for supporting personalized ads on various platforms for all types of channels Most engineering aspects of these products are already addressed in this work

    Technology Inputs:

Imported Equipment/Spare Parts:
Equipment/Spare Parts Year ITC-HS Code
NA
Indigenous Equipment/Spare Parts:
Equipment/Spare Parts Year ITC-HS Code
Workstation NA
Mobile Workstation NA
High End Desktops NA
Network Attached Storage NA
TV Broadcast Video Acquisition Accessories NA
External Storage Hard Disks NA
Imported Raw Materials:
Raw Materials Year ITC-HS Code
NA
Indigenous Raw Materials:
Raw Materials Year ITC-HS Code
NA
Existing R&D Facilities used:
Facilities Year ITC-HS Code
NA

   Patents & Publications:

Patents:
Filed Patents (No.) Granted Patents (No.) Year
0 0 NA
Publications:
Submitted (No.) Published (No.) Year
2 12 NA

    Commercialization Potential:

Who are the Potential Licensees? TCS Innovation Labs TATA Power SED EMMC GoI
What commercially available products address
the same problem?
Company Product Problem Addressed
Would you like to develop this invention further with
corporate research support?
Yes
Would you be interested in participating in cluster based
programs for commercialization research or business
planning for your invention?
Yes
      Submitted by: Research and Development Section Date of Submission: 7-8-2020



Chat Room      Write Review     Talk to Experts


THE VISION
THE MISSION
ABOUT I-STEM
It has always been the basic tenet of the Government of India, in generously funding R&D efforts at academic institutions over the years, that facilities established through such support be made available to those needing them and qualified to make use of them for their own research work
read more >>

However, this was never easy or straightforward for, among other reasons, there was no ready source of information of what facility was available and where. Thanks to the Web, it is much easier today to have a national and regional “inventory of resources”, so as to match users with the resources they need, and to do all this in an efficient and transparent manner.

This can lead to a leap in R&D productivity and greatly enhance the effectiveness of public investment. This is the motivation behind I-STEM.
read less <<
Visitor Hit Counter
Hosted at Indian Institute of Science
Copyright © 2024 I-STEM. All rights reserved.
Audited by: STQC Bengaluru.