CSC News

October 29, 2014

A Virtual Whodunit

NC State Engineering | Pack Points
Crime scenes are fleeting.
Police and first responders are trained to disturb as little as possible at the scene of a homicide or other violent crime. But no one’s perfect, and eventually the scene must be released. Homes must be occupied. Businesses must reopen. Traffic on a busy street must flow once more.
But what if you could preserve that space virtually, so that investigators and the specialists with whom they collaborate could return to the scene of the crime during the investigation?
That is the aim of IC-Crime, a multidisciplinary project led by faculty from NC State’s College of Textiles working alongside colleagues from the Department of Computer Science and the College of Design to give real law enforcement a tool that might fit on the popular television drama “CSI.”
IC-Crime’s development is part of NC State’s Digital Games Research Center (DGRC), housed in the Department of Computer Science on Centennial Campus. Work on the design of the IC-Crime software is overseen by Dr. Michael Young, professor of computer science, director of DGRC and a co-principal investigator on the project. Dr. David Hinks, Cone Mills Professor of Textile Chemistry and interim dean of the College of Textiles, is the principal investigator.
“The project has been a fantastic collaborative effort amongst computer scientists, industrial designers, management technology experts and textile chemists, all of whom are interested in contributing their expertise to the field of forensic science, which is ultimately applying science to the law and protecting the innocent as well as effectively convicting or identifying guilty parties,” Hinks said.
On the (virtual) scene
In a world imagined by IC-Crime, an expert on blood spatter would sit down at her computer and access a secure server online. Investigators faced with a particularly vexing homicide case need her help.
She starts as an avatar walking into a virtual lobby. From that lobby, she chooses from several rooms in which to enter. Each room is a crime scene in another state or even another part of the world that she can see without leaving her desk.
Once inside the scene, she will be able to move around the room and look at things from different angles. Just as in a real crime scene, this room is filled with markers. By clicking on these markers, she is able to pull up additional data points like high-resolution photographs, information on shell casings or lab reports.
The homicide detective who has asked for her help is also in the room, represented by an avatar of his own. They are able to speak with each other within the virtual environment and even trade views of the room.
The work done to create this virtual environment was done by crime scene investigators as the scene was processed. After CSIs dusted for fingerprints and scoured the room for fibers, they also set up a 3D scanner on a tripod that took a digital scan of the room.
The kind of traditional paper trail that is part of an investigation has been scanned and added to the virtual setting, so the avatars can quickly pull them up by clicking on those markers.
And it doesn’t stop there: The infinite power of the digital realm allows access to large databases. One example is an unprecedented database of commercially used dyes being assembled in NC State’s College of Textiles under Hinks’ leadership. Using that database to match dyed materials taken as evidence would allow detectives a useful tool to exclude or focus in on a suspect. The IC-Crime environment would make that tool available to all the collaborators in an investigation.
The software is integrated into the Unity3d computer game engine.
Unity provides a powerful 3D software environment to build upon, Young said, and allows the team to deploy the software remotely across a number of platforms.
Collaborators can log into the Unity environment via a Web browser without having to download any software onto their computer.
Chasing down leads
The IC-Crime team at NC State has worked with the NC Program for Forensic Sciences, the North Carolina State Bureau of Investigation, the Wake City-County Bureau of Identification, and the North Carolina Justice Academy, among others.
Detectives from the Fayetteville Police Department invited the NC State research team to use a house in Fayetteville to set up a mock crime scene and scan it. The team staged a similar scene in the main hall of the Park Alumni Center on NC State’s Centennial Campus. Young played the unfortunate deceased during that simulation, a task that required lying motionless for several hours.
Collaborators from Arizona State University and Indiana University are conducting social science research that examines how the software affects the perceptions and understanding of the crime scene by the investigators who use it.
Tim Buie, an associate professor in industrial design, works with Young to create the visual elements within IC-Crime’s interface. Dr. Roger Azevedo, a professor in NC State’s Department of Psychology, is using tools including biometric data to study how a typical user interacts with IC-Crime and whether it can be made more user-friendly.
Young says working with artists, sociologists and psychologists is typical when building the kind of games being developed by the DGRC. These so-called “serious games” take core games technologies traditionally used for entertainment and use them to address a broad range of issues, including training, education, security, healthcare or social interaction.
“Games by their nature are multidisciplinary. They define an intimate relationship that sits at the boundary between technology and a player,” Young said. “It’s very natural that there’s an engineering component to the research, but also a strong humanist element.”
All that work has led to a system that is ready to go to the next phase, one that will transform it from a research tool to an investigatory one.
The three-year National Science Foundation grant that started the project will end in January. While Young and Hinks are seeking additional funding to continue research, the first version of the IC-Crime software is nearly complete.
Hinks says the NC State team is looking into proposals that include industry and university partnerships, and there has been interest from software companies in the United States and elsewhere.
Law enforcement agencies that want to use IC-Crime will need a digital scanner, a secure server to store the information and the flexibility to scan a crime scene. Scanners are expensive, but Young says that advances are being made — software that works with the Microsoft Kinect can make the motion-sensing device into the kind of digital scanner that is needed, and Google is working on smartphones that have 3D data acquisition capability.
In order for the system to be practical, scanning of crime scenes must be more efficient. It takes about 15 minutes for a digital scanner to take a 360-degree image of a room and because of laser shadows created by obstacles in the room, the room must be scanned more than once from different vantage points. Because of the impact the scanning laser has on the human eye, the room must be vacant.
The hope is that industry will perfect the IC-Crime system and then take it to crime labs that need the help.
“The work isn’t done,” Hinks said. “In many cases, we’ve done a lot of the groundwork, the basic work for IC-Crime. But there are many future levels that could be developed and should be developed.”
Closing the case
Hinks says he received an encouraging insight into IC-Crime’s potential uses when discussing the idea with a Fayetteville Police detective who said he could see utilizing the technology during a suspect interrogation.
“So that’s your version of the night’s events?” a detective might ask a suspect. “Watch this digital simulation of exactly what the crime scene looked like that night. It shows that what you are telling me is impossible. Would you like to change your story now?”
“For me, that was quite illuminating,” Hinks said.
Whether IC-Crime technology goes beyond the precinct house and into the courtroom remains to be seen. The benefits seem obvious: While jurors are sometimes taken to a crime scene, the passing of time means what they find is always different than what prosecutors and defense attorneys have described. With IC-Crime, they could see what that scene looked like the night of a murder.
But is that what prosecutors, defense attorneys, judges and even investigators want?
Hinks points out that technology, including digital photography, is slow to make its way into the courtroom as defense attorneys find ways to poke holes in the new methods’ accuracy and cast doubt in jurors’ minds.
“There’s tremendous benefit to be able to bring the crime scene to the jury as it was, but it’s still a simulation and that will be challenged in the courtroom, no doubt,” Hinks said. “My anticipation is that there will be some states that would be very reluctant to go down that road, and then there will be some states that will be quite open.”
And investigators and prosecutors worry about what they refer to as “the CSI effect.” The whiz-bang technology deployed on popular crime investigation television shows isn’t always available for real-life law enforcement agencies, and the lack of the kind of technology jurors are familiar with from sitting on their couch might make it hard for them to convict when sitting in a jury box instead.
Other potential applications go beyond the homicide beat.
Crime scene photographers must learn what to shoot on the scene and how to shoot it. Staging crime scenes can be time-consuming and expensive. A virtual scene could provide a cheaper, easier training ground.
New prosecutors and defense attorneys need to know what to look for at a crime scene. So do investigators. The training potential for IC-Crime could be just as strong as its potential use in investigations.
Hinks even sees possible military applications as a tool to reconstruct a battlefield and determine what happened. Young mentions the US Department of Defense as a potential source for a next round of funding.
As Hinks says, “the platform that we are building has tentacles that can run into different arenas.”
It’s a project that provides a different perspective on the usefulness and impact of computer games.
“There’s a breadth of new capabilities that game technologies enable that go well beyond entertainment,” Young said. 

Return To News Homepage