DeepCrowd: A Crowd-assisted Deep Learning-based Disaster Scene Assessment System with Active Human-AI Interactions

Time Frame

2021-2023

Total Funding to Date

$499,786.00

Investigator

  • Dong Wang

Full title: CHS: Small: DeepCrowd: A Crowd-assisted Deep Learning-based Disaster Scene Assessment System with Active Human-AI Interactions

This project addresses the application of AI to disaster scene assessment (DSA). AI currently has limited success with DSA; this project investigates the problem of troubleshooting, tuning, and eventually improving AI algorithms by integrating human intelligence with machine learning. 

The end result will be a DeepCrowd framework that can be used to guide the design, development, and implementation of future AI applications where human intelligence obtained from the crowd is integrated with AI learning models.

DeepCrowd

Funding Agencies

  • National Science Foundation, 2021 – $499,786.00