Video-tracking evidence is admissible in California

What is video tracking? A look at People v. Tran and how video-tracking could be helpful in civil cases

2022 January

Video-tracking enhances and combines more than one video image from different sources into one chronological and comprehensive image concerning an event that may involve a large number of people or an accident or assault. People v. Tran (2020) 50 Cal.App.5th 171, is the first California authority to hold that video-tracking and the accompanying explanatory testimony are admissible in evidence.

Tran was convicted of being involved in assaulting and seriously injuring M.C., who is now paralyzed from his shoulders down, during a public brawl in San Diego. M.C. was injured when he was body-slammed head-first onto the pavement. No one positively identified Tran, or anyone else, as the one who slammed M.C. onto the ground and caused his paralysis, although one of the videos showed Tran almost sitting on M.C.’s paralyzed body and punching him at least five times in the face.

Grant Fredericks, the prosecution’s certified forensic-video expert witness, used five different videos of the melee obtained from nearby hotels and restaurants and a bystander’s cell phone to chronologically sequence what had happened that evening. Fredericks uses video synchronization, image-comparison investigations, aspect-ratio calibration, color correction, motion tracking, and image authentication to create video- tracking. He testified that he focuses on preserving the integrity of a video and that the images are untouched. He said he employs a variety of tools to “interrogate” each video file and analyze compression levels, frame rates, and pixel matrixes. Fredericks testified that the procedures he described are “currently [used by] several hundred certified analysts and technicians, all of whom have been trained in this process.”

Fredericks said that by synchronizing minute details like clothing patterns and the location of one’s feet, he is able to track an individual through multiple videos, even though that individual’s image might have otherwise been too blurry when simply viewing a specific moment in one video. Fredericks assigned a color arrow to an individual when he could validate its accuracy with forensic certainty. He acknowledged that “where I have assigned an arrow, my opinion is it is the same person.” Fredericks cautioned that the lack of an arrow “doesn’t mean the person is not there,” but rather, “just means that I can’t prove to a certainty that that is the person.” Fredericks assigned a blue arrow to Tran, who had particular dress and hair characteristics, and showed that the person designated by the blue arrow appeared in certain numbered frames, including when he repeatedly punched M.C.

Fredericks explained that the average person who separately views multiple videos may not appreciate how and when they capture the same scene at the same time. He said that people have difficulty understanding the significance of blurring in a video, which could represent fast motion. According to Fredericks, jurors also might fail to appreciate how infrared impacts various colors – and more specifically, the colors of a person’s clothing – causing a person to appear to be wearing different clothing in different videos of the same event.

Fredericks did not offer an opinion about the nature of M.C.’s fall, what injuries were caused or who threw M.C. to the ground.

The trial court allowed Fredericks’s testimony and the video-tracking chronology after conducting a hearing under Evidence Code section 352 concerning whether the probative value of that evidence substantially outweighed its prejudicial impact.

Tran appealed his conviction for assault by means likely to produce great bodily injury and mayhem, contending, inter alia, that the trial court erroneously admitted into evidence videos used by Fredericks.

Although Tran did not specifically argue that Fredericks’s enhancement of the videos should have been examined under Kelly-Frye principles and the appellate court found that he had waived that contention (id. at p. 178), it went on to observe that the trial court had ruled that there was no need to apply Kelly-Frye because “the methods and procedures described by the expert are not a novel, scientific testing method or procedure.” (Id. at p. 182.) The appellate court agreed with the trial court, observing that neither Fredericks’s testimony nor the videos were subject to Kelly-Frye. “Like the animated video in (People v. Duenas (2012) 55 Cal.4th 1), Fredericks’s work is a form of computer animation analogous to “charts or diagrams” used in other ‘classic forms of demonstrative evidence.’ (Duenas, supra, at p. 20.) Fredericks did not alter the underlying surveillance videos, except to enhance their quality and correct pixel ratios. In fact, he aimed to preserve the integrity of the videos. In addition, Fredericks testified that his technique for tracking and sequencing videos involved both common sense and long-standing tools routinely used for image comparison.” (Id. at pp. 187-188.)

The appellate court found no problem with the trial court’s decision to admit the videos and Fredericks’s testimony and affirmed Tran’s conviction. It explained:

Based on the record before us, we conclude that the trial court did not abuse its discretion in admitting the videos or Fredericks’s testimony about the videos. … Fredericks’s testimony helped the jury to identify what portions of the video evidence required closer examination and to interpret some of the information conveyed by the video evidence. His expertise was necessary for the jury to accurately evaluate the videos to appreciate who and what they were watching as well as the chronology and relationship between each video. As the trial court noted, the individuals in the videos were very difficult to track ‘even with the benefit of [Fredericks’s] colored indicators,’ and ‘impossible’ to track without his expert assistance.

Although jurors ordinarily may be capable of watching a surveillance video and understanding what they see without expert help, the scene captured by the multiple videos was especially challenging. As such, Fredericks’s assistance in this case was critical because multiple surveillance videos depicted a moving melee, at night, with at least a dozen bodies interacting on a crowded street. Fredericks’s testimony and sequenced videos were necessary to aid the jury’s understanding of how the different videos captured the same action from different angles, thus minimizing confusion that can result from contrasting video perspectives, contradictory descriptions, or unreliable eyewitnesses. Indeed, Fredericks’s synchronization demonstratives and tracking videos were critical for the jury to make sense of the various videos that differed by way of time stamps, camera angles, dimensions, lighting, resolution, pixel ratios, and quality. Further, as the crime occurred at night and the lighting of the streets and in the videos was not uniform, Fredericks’s testimony and enhancement of the videos aided the jury in identifying the various participants in the melee.

(Id. at pp. 188-189.)

It is not difficult to see how this kind of video-tracking technology could be helpful in all manner of civil cases, for both plaintiffs and defendants. Cases involving traffic accidents, assault, and excessive use of force by the police would seem to be excellent candidates. The utility of this technology underscores the need for counsel to identify percipient witnesses who might have recorded an incident, and to be aware of possible sources for surveillance video.