Page 239 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

Use Case 14: Remote Visual Tracking Using Semantic Scene

Understanding and GenAI-based Predictive Rendering Under
Disaster Situation Change 4.2-Climate

Organization: Tata Consultancy Services

Country: India

Contact Person:

Primary: Abhijan Bhattacharyya, Abhijan.bhattacharyya@ tcs .com
Secondary: Ashis Sau, ashis.sau@ tcs .com
Additional Contacts: Suraj Mahato, surajkumar.mahato@ tcs .com

1 Use Case Summary Table

Item Details
Category Climate Change/Natural Disaster

Problem Addressed In case of massive disaster (like flood) over a vast area, the terrestrial
cellular connectivity may get destroyed. This jeopardizes the ability of
remote instantaneous visual tracking of the disaster site by the central
control room. Even sending HD images or videos over Geostationary
(GEO) satellite based communication channel leads to high latency
and suffers from lack of bandwidth in practice. This leads to delayed
comprehensive disaster response strategy.
In this context we propose a low-bandwidth consuming system based
on the principles of semantic communication, coupled with GenAI
based rendering to transfer ‘semantically matching’ live visualization
of the remote disaster.
It may be noted that although Low Earth Orbit (LEO) satellites are prom-
ising for providing broad-band internet service for future, but back-haul
on LEO has several operational, privacy and security issues due to their
low-footprint and mobility. This also gives rise to many challenges due
to the handover requirements and inter-satellite link congestion. So,
GEO satellites will continue to be a reliable backhaul for critical utility
services utilizing non-terrestrial network (NTN) [1].

203