Case Study: 3D Motion Capture to Determine Number of Killers in Italian Kitchen

The woman’s body was found at her apartment entryway in the kitchen, mired in a pool of blood. She had defensive wounds covering her hands and trauma from a bladed weapon and a blunt weapon.

The autopsy ruled it as a homicide that was likely committed by two attackers—but the homicide investigation focused on a single killer, based on circumstance and motive outside the scene.

Motion capture technology involving a 3D reconstruction of the brutal murder in southern Italy helped solve the mystery, according to a new case study published in the Journal of Forensic Sciences.

“Our 3D model allowed us to experimentally reproduce the murder and compare various dynamics,” write the authors, from the University “Magna Graecia” of Catanzaro. “The model was used as evidence in court (…)”

The autopsy showed the wounds were all inflicted while the victim while still alive.

Eleven lesions were made with a knife, three of which were penetrating (one that lacerated the left lung, and two on the right side of the body that pierced the liver).

Twenty blunt wounds caused significant cranial fractures—and brain trauma was the cause of death, according to the study.

The team put the totality of the evidence in and around the body into a 3D modeling system.

The objects in all the rooms in the apartment were examined, and all were cataloged and photographed on a planimetry schematic. The bloodstain patterns found in the kitchen and entryway were added into the model.

The injuries of the victim cataloged from the autopsy were numbered, photographed, measured and triangulated, too.

The suspect identified by police detectives also had abrasions on his face, which were added into the mix.

A search of the suspect’s home turned up clothes with biological traces of the victim on them, which were also incorporated into the reconstruction, according to the paper.

Two actors were then incorporated into a 3D motion capture sequence using infrared OptiTrack cameras, which can capture 100 frames per second. The two actors had the same height and weight and were tracked to capture the movements thought to have been taken.

The simulated attack was GPU rendered using a Supermicro system with multiprocessor Xeon quad and NIVIDIA Quadro Plex systems.

The reconstructions then became a system of trial-and-error, to see which best fit the evidentiary pattern.

“Different sequences were progressively excluded because they were incompatible with the pattern, direction, or lethality of the blows, and the reconstruction led the investigators to a compatible order of events,” write the authors.

The simulation hit on a working pattern: knife in the left hand, and the blunt object in the dominant right hand of the offender.

The murder occurred in three phases: an initial attack, a struggle, with objects hitting the ground around the scene, and the death blows made by the blunt weapon as the victim tried to defend herself. The head injuries were first made on the occipital region, and then progressively toward the right temporo-parietal region of the brain, leading to the woman’s death.

“A 3D model allowed us to prove that a single offender could kill the victim with two weapons at the same time and that no other killers were involved,” they write. “This has given great credibility to the hypothesized dynamics of the murder, and allowed the public prosecutor to understand the phases of the struggle and the brutality of the blows inflicted.”

Limitations included the potential for bias with rendering of the crime, and also the potential cost for the computing and camera equipment.

Overall, however, the case study could be an example for cases in years to come, conclude the eight Italian scientists.

“This case illustrates how the correct use of 3D technology could become an essential instrument in crime investigation over the next few years,” they conclude.

By Forensic Magazine.

Leave a Reply

Your email address will not be published. Required fields are marked *