Morgan Smith

Morgan Smith

Proper Digital Discovery, Part V: Reality Capture

In our last post, we examined the use of drones in developing the 3D model. In this post, we will examine some of the best methods and tools for preserving & maintaining evidence for use in litigation: reality capture. Reality Capture software and hardware — applications and systems that specialize in capturing and digitizing the reality of the world around us — increasingly are becoming more prevalent within today’s available options for professional trades. Not too long ago, a tape measure was an expert’s best friend, but things have changed dramatically.

Part Five:

Combining Drone Video and Laser Scans

These days, experts who analyze the physical world (i.e. architects, builders, police officers, litigation experts, etc.) still grab their tapes when they’re headed to a job site; however now, their primary tools are often technical marvels such as UAVs (Unmanned Aerial Vehicles, aka, drones) coupled with LiDAR and/or structured light laser scanners. Such technologies can be paired together within very sophisticated and expensive software programs, producing highly accurate, 3D digital models. These “captured” references/representations of reality can be reviewed, measured, and analyzed at a later date, thus archiving site conditions in an intuitive, three-dimensional format for foundational purposes. Often while processing video/photography footage from drone flights, in conjunction with 3D laser scans, it can become necessary to manually align aspects of the coverage that do not fully sync up automatically. This alignment process can be quite tedious if there is not sufficient video, photos, and/or LiDAR scans. A fair amount of overlap is needed within datasets in order to properly align all data together into one dense, 3D point cloud. If/when manual alignment is needed, CAD operators will pick points (i.e. at least 3 features within each image(s), video frame(s) and/or LiDAR scan(s) to be aligned), and these points of reference help the software determine how these datatypes should shift accordingly. Once all datapoints are in sync, the CAD operator will be able to produce a very realistic 3D mesh that will be used as the basis for diagrams, animations and supporting materials for expert witnesses.

Below is a video showing a combination of aligned LiDAR point clouds and drone video/photography footage; this will be used to create a 3D model.

01. LiDAR + Video = Accuracy

High resolution 3D model generated from drone video & LiDAR

The combination of both a laser scan and drone video/photography is often the best and most efficient way to obtain a realistic and highly accurate recreation of a case’s scene. However, it does have its limitations. Notably, the computer is interpreting how to align all footage captured; so, it’s not perfect every time. Occasionally you’ll get a combination of imperfect alignments and/or bizarre 3D artifacts (e.g. phantom/floating, unsubstantiated, “blobby” 3D objects) created erroneously while algorithms process the equivalent of a 3D or 4D puzzle. The software is doing its best to make sense of a lot of varying datasets.

As you can see from the sample above, using drone photography can allow for the development of convincing overview diagrams that can look really good at a distance. However, often, if only a typical level of detail is collected by drone video/photography, the resolution at a pedestrian view can be hit-or-miss; the quality can diminish a bit into something less than photoreal.

To offset some of this loss, often supplemental efforts are needed: perhaps more pedestrian photographs are needed (taken with an SLR camera), overshooting with more LiDAR scans and/or using such scans to manually 3D model scene elements can help to refine geometries. This will lead to more realistic final products.

Basically, if you want to recreate a scene at eye level, that looks photo-realistic, nothing beats having designers use underlying LiDAR and/or drone imagery as a basis to create accurate 3D models. A designer can use these detailed datasets as background, and then model foreground (near elements so that eye-level views of a scene will look realistic) based on accurate foundation (See prior post on how a 3D modeler uses a laser scan to create a 3D model). For the most part, only a small portion of the scene will need to be enhanced by hand through a 3D modeler’s artistry. The balance of LiDAR laser scans and drone imagery will effectively provide distant context and dimensional accuracy for the scene.

Below is an image from a case where we 3D modeled buildings (close to the road) as well as the road. We then used scan images for the background which was less important.

3D models using foundational data from LiDAR scan, with top view overlay

02. Adding Evidence Photographs Into 3D Models

One of the more interesting technologies, now available, is the ability to use photogrammetry software – capable of compiling thousands of photographs, videos, drone imagery and/or LiDAR scans – while ultimately being able to add in evidence photographs and/or video, matching it to its original location(s) within a scene.

For instances, let us say you have a 3D model created from a highly accurate laser scan. By using such photogrammetry software, a CAD operator is able to automatically or manually align photographs into their original locations from where the evidence was collected. Long after the incident occurred, it is possible to definitively tie evidence to a 3D LiDAR/drone scan, so that attorneys’ experts can make observations and provide analysis based on position-located evidence.

Every camera lens has a different effect on what we see in a photograph. These differences can make photogrammetry and/or photo-matching a view within a 3D scene challenging. Trying to match the exact vantage point can be challenging if not enough context is present. Within an earlier post, you may recall the importance to always ask for electronic data that is in its native format. Native metadata, from a digital photograph, will indicate a camera’s attributes, including the specific camera and lens used for that specific evidence, as well as a host of all sorts of other potentially useful information. Provided this information, photogrammetry and/or 3D modeling software will allow you to dial-in the precise camera/lens configuration. This specificity will allow CAD operators and/or computer applications to recreate the scene more easily. Assuming all parameters are in alignment, original photographic evidence, from the actual scene, will usually sync up and match a fuller, reality captured context.

Previous
Next

Cogent Legal can create lifelike 3D models of any object or scene

At this point, you may be wondering:

Why should I care if the exhibit photographs are placed within the scene of my case properly?

Imagine for a second that you have many photos of evidence spread throughout your incident’s scene. If this scenario is true, then being able to import all such photos accurately into one three-dimensional landscape creates a historical account, in a sense preserving the integrity of the evidence within its vicinity. Over time often scenes of an incident will change due to repairs or modifications following an incident. In essence, linking past photographs to new drone/photography and/or LiDAR footage, creates the equivalent opportunity as if your own experts were present at your scene, moments after the incident took place.

Previous
Next

LiDAR scanners calculate exact distances on thousands of points to achieve 100% accuracy

Once such a 3D scene is created, not only can a narrative animation be crafted using that model, but also very accurate diagrams, including measurements from this virtual scene can be annotated with little effort. The team at Cogent Legal are experts in this field. Give us a call – we’d love an opportunity to capture and preserve the most important details of your case – the earlier the better.

Diagrams based on LiDAR data can further simplify the message

Cogent makes it easy.

Laser scans and drones have replaced the measuring tape of yesterday. Understanding the power and limitations of this changing technology is a key aspect of modern litigation. We are able to create accurately scaled diagrams, such as floor plans, accident reconstructions, maps, and so on, directly from 3D data collected by our team. We work with you and your experts to build 3D models that allow audiences to connect with your case’s details from any angle, helping everyone get on the same page and develop the most effective arguments. Call Cogent Legal today and make sure your case is handled the right way—from the very start.

Evidence collection & preservation - by Cogent Legal

Share this article:

Share on facebook
Share on twitter
Share on linkedin
Share on print
Share on email

Set yourself up for the best case scenario.

Let's get in touch