Five questions to answer before buying a video redaction solution
By Sean Varah
CEO/Founder, MotionDSP Inc.
Over the past five years of shipping a video redaction product, my team at MotionDSP has met with numerous police departments, district attorneys, universities and body camera manufacturers. We’ve learned a lot from those conversations and thought we would share a list of the 5 key questions customers should ask before purchasing a video redaction solution for FOIA requests.
1. Does it work with body camera footage?
Some companies promise that their redaction solution can automatically track faces so the user doesn’t have to perform manual frame-by-frame redaction. However, if you look at some of vendors’ demo videos, they often use video footage from a fixed camera rather than a body camera. For example, Microsoft just announced a developer tool for redaction, and its demo video is two people dancing in front of a tripod-mounted camera.
Body cameras are far from stationary. They are worn by humans who walk and run. Tracking moving objects from a moving camera is a far greater computer vision challenge than tracking from a fixed camera. Software with poor tracking will force you to revert to manual redaction for every frame.
2. Is it easy to use for inexperienced users?
Many video redaction solutions were designed for highly trained forensic video analysts to use. What we’ve heard from most of our customers is that they are using interns, new hires or temporary consultants to produce redacted videos for FOIA requests and can’t afford to spend weeks training these employees. They need software that can be learned in hours. For example, some departments consider purchasing video editing software to use for redaction, which may have a cost advantage, but it requires their employees to learn to complex software before they can learn to redact video, adding days or weeks to training.
3. Is it frame accurate?
As we wrote discussed in our article, “Don’t believe the hype behind automated redaction,” automated redaction sounds great until you consider the risk of the software failing for even a single frame of video. If you expose an innocent person’s face, your liability goes through the roof. To make sure you have every frame of video redacted properly, your software needs to be able to pause accurately on any frame of the video. The major cloud redaction solutions use standard web video technology that isn’t able to do this (internet video was designed for Netflix and YouTube, not video redaction), and so you can never be sure if your redaction is accurate.
4. Does it work with other vendors’ video files?
Police departments investing in a redaction solution quickly discover that they need something that can work with any sort of video, not just body cameras. Video from in-car video systems, mobile phones and fixed surveillance systems also needs redaction. Many cloud-based solutions are only compatible with the cameras sold by those vendors, so while you may have a redaction solution for body cam footage, you’ll have to invest additional dollars to redact all your other video files.
5. Does it accurately record and report the redactions you make to the video?
Transparency in law enforcement is crucial, especially in court. After talking to many district attorneys, they mentioned that having a redaction report that details exactly where blurs or audio redactions take place is crucial for use in court. We had this feature in our video enhancement software, Ikena Forensic, but also added it to Ikena Spotlight in our most recent release due to it being a must-have feature for many customers.
Choosing a body camera and redaction solution is a complex undertaking. We’ve worked with the law enforcement community for over 8 years, and we’d be happy to help anyone in the community with their technical questions before they make a major purchase. Reach out to us at firstname.lastname@example.org.
MotionDSP has delivered cutting-edge video solutions to law enforcement customers since 2008. Ikena Spotlight, our video redaction solution, was developed in 2010 after hearing the London Met Police describe their challenge of manually blurring out civilian’s faces from CCTV videos.