Technical Report of the Video Event Reconstruction and Analysis (VERA) System - Shooter Localization, Models, Interface, and Beyond

05/26/2019
by   Junwei Liang, et al.
2

Every minute, hundreds of hours of video are uploaded to social media sites and the Internet from around the world. This material creates a visual record of the experiences of a significant percentage of humanity and can help illuminate how we live in the present moment. When properly analyzed, this video can also help analysts to reconstruct events of interest, including war crimes, human rights violations, and terrorist acts. In this technical report, we describe a newly developed tool, the Video Event Reconstruction and Analysis (VERA) system, that enables the localization of a shooter from just a few videos that include the sound of gunshots using established machine learning techniques and straightforward physics models. This tool relies on other tools we have already developed including video synchronization and geolocation to order unstructured videos lacking metadata over time and space, and sound detection algorithms. Both this gunshot localization tool and the previous systems it incorporates are run through a web interface that enables human-in-the-loop verification to ensure accurate estimations. To demonstrate the efficacy of this suite of tools, we present the results of estimating the shooter's location of the Las Vegas Shooting in 2017 and show that VERA accurately predicts the shooter's location using only the first few gunshots. We then point out future directions that can help improve the system and further reduce unnecessary human labor in the process. All relevant source code, including the web interface and machine learning models, are freely available in Github. We do so in the hope that this tool can be used by anyone who needs it to protect and promote human rights and public safety. We also hope that researchers and software developers will be inspired to improve and expand this system moving forward to better meet the needs of public safety and human rights.

READ FULL TEXT

page 1

page 2

page 4

page 5

research
05/26/2019

Technical Report of the DAISY System -- Shooter Localization, Models, Interface, and Beyond

Nowadays a huge number of user-generated videos are uploaded to social m...
research
11/02/2017

Framework for evaluation of sound event detection in web videos

The largest source of sound events is web videos. Most videos lack sound...
research
01/25/2018

SocialML: machine learning for social media video creators

In the recent years, social media have become one of the main places whe...
research
06/26/2023

INDEXITY: a web-based collaborative tool for medical video annotation

This technical report presents Indexity 1.4.0, a web-based tool designed...
research
07/05/2017

Web Video in Numbers - An Analysis of Web-Video Metadata

Web video is often used as a source of data in various fields of study. ...
research
03/23/2022

Muscle Vision: Real Time Keypoint Based Pose Classification of Physical Exercises

Recent advances in machine learning technology have enabled highly porta...
research
06/25/2021

Usage-based Summaries of Learning Videos

Much of the delivery of University education is now by synchronous or as...

Please sign up or login with your details

Forgot password? Click here to reset