How Virtual Crime Scenes Work

The first virtual crime scenes were created in computer models. Investigators took careful measurements at a crime scene, including the dimensions of the scene itself and the relationship between all the objects inside the scene. For example, at a murder scene in a house, an investigator measured the dimensions of the room, the position of the body and the location of any other related information, like weapons or blood. Often, investigators used laser distance meters or a total station (a surveyor's tool) to ensure accurate measurements. Crime photographers took pictures of the scene from multiple angles. After the initial crime scene investigation, the investigator returned to the police station.

At the station, a computer operator took the measurements gathered by the investigators and entered them into a computer program. The program then rendered a three-dimensional space using that information.

Some programs required human intervention to make the spaces accurate -- one particular approach required a programmer to adjust the height of a room manually. The programmer used a Computer Assisted Design (CAD) program to adjust the computer model, create furniture or insert the figure of a body. Some early programs also included a library of common objects, like file cabinets or desks, to make the process a little faster and, therefore, less expensive.

Time and money were the two biggest obstacles to making virtual crime scenes a useful tool. The early programs needed a highly skilled computer programmer or graphics designer to make the environments convincing enough to be useful and immersive -- in other words, they didn't give the viewer a sense of presence within the virtual scene. Police forces needed software that had a much less steep learning curve and automated as much of the process as possible.

Today, several software companies offer products that automatically build rooms based on either basic measurements or a two-dimensional room plan. Most of these programs also include extensive object and texture (overlay graphic) libraries, allowing the user to customize the virtual environment and make it as accurate a representation of the real crime scene as possible. Most of these programs can even include the flight path of a bullet (as previously determined by investigators using standard forensics techniques).

The advantage to the improved virtual environment is that once it's rendered, users can now view the scene from any angle, including from vantage points that are difficult to access in the real world. By looking at the scene from different perspectives, investigators have more information that can help them accurately determine the course of events at a crime scene. Some programs also allow a user to add in animations, allowing investigators to view their theories in action.

If the program supports video output like a head-mounted display, the user can experience a strong sense of immersion, which can help witnesses recall more details or allow investigators to check lines of sight.

But modern virtually rendered environments have disadvantages. The biggest disadvantage is the need for an employee who has enough technical savvy to work with the program -- even with programs that automatically build environments, a user often has to make adjustments. Another disadvantage is that virtual environments are rarely perfect copies of the real crime scene. As a result, investigators or witnesses may become distracted by the imperfect simulation and miss important details.

In the next section, we'll look at a different method of constructing a virtual crime scene.

Some photogrammetry systems use a special camera mounted on a tripod that can rotate the camera's view in 360 degrees. The camera takes a series of pictures and sends them to a computer that has proprietary software designed to combine the photos together to make a seamless photograph of the environment. Users can change the point of view for the photo, turning in any direction.

These photogrammerty systems are much more efficient than a human photographer snapping photos from every viable angle. However, since it's on a stationary tripod, the camera won't capture any areas in an environment where the view is obscured from the tripod's position. Investigators have to move the tripod and take another series of photos to capture that perspective.

In many ways, photogrammetry has significant advantages over creating a graphical representation of a crime scene. For one thing, you don't need to be a skilled programmer or graphics designer to use the hardware and software. The machinery and computer programs do all the work for you, creating comprehensive photos in a few minutes. Another advantage is that it's not just a representation of the crime scene -- it's a series of actual photographs. Viewers won't be distracted by an unrealistic object model or a poorly chosen texture.

But there are some distinct disadvantages. Unlike a virtual model of a crime scene, you can't view a room from any angle -- you can only look around from where the investigators set the tripod for the pictures. Unless the investigators put the tripod in multiple places within the same space, you're stuck viewing a room from one fixed point. Photographs also aren't very good at accurately showing the depth of a view -- objects may seem to be closer or farther away than they should be.

Some photogrammetry software packages are very affordable, but the hardware tends to be expensive. For some police departments, the expense is justified because the software doesn't require someone with special training to use it. Some programs even allow users to use a normal digital camera instead of a specialized photogrammetry device. These programs require the photographer to shoot the same scene from multiple angles, then use the computer software to identify identical points in each photo. The software can then create a three-dimensional model by triangulating the identical points in each photo. This solution is much more affordable, but is less comprehensive than advanced photogrammetry techniques and requires more man hours to use.

In the next section, we'll look at some applications for VR crime scene technology.

­

Why are police departments investing thousands of dollars in this sort of technology in the first place? For years, investigators worked from sketches, photographs and scale models to solve crimes. Why bring virtual environments into the process?

The most important reason is that investigators can preserve a facsimile of the crime scene indefinitely. Some crime scenes are in areas that receive a lot of public traffic and can't be cordoned off for an extended length of time -- the crimes scene clean up crew must return the scene to its normal state. By creating a virtual copy of the scene, investigators can revisit it whenever they want without inconveniencing anyone.

The applications for virtual environments in crime investigations include:

As companies develop user-friendly applications, we'll likely see police use VR technology more often. Several forces have expressed interest in using the technology to explore cold cases, possibly leading to an arrest decades after the initial crime. Many others are already using the technology to share investigation results with officers from different forces.

To learn more about virtual crime scene investigation and related topics, check out the links on the next page.