<img height="1" width="1" style="display:none;" alt="" src="https://dc.ads.linkedin.com/collect/?pid=49414&amp;fmt=gif">
 

I’d like to learn more about Campaign Creators!

Automatic Face Redaction – A Solution to Many Privacy Concerns

There are privacy and other concerns surrounding face detection and recognition technology. Can automatic face redaction be a possible solution?
by Muhammad Nabeel Ali Updated on February 22, 2022

Privacy concerns have been predominant and prevalent when it comes to handling enormous amounts of data. One such scenario where this phenomenon occurs is when we work with computer vision technologies such as face detection.

Our blog will be discussing matters revolving around privacy concerns with face detection and recognition technology. Later on, we will present a possible solution for the issue at hand – face redaction and how to use VIDIZMO and AI to redact faces.

Face Detection And Face Recognition

Face detection is an incredible development in the field of artificial intelligence and computer vision. However, it has been considered questionable in many regards. Facial recognition, which allows identification of the faces being detected, also adds to the list of concerns amongst people.

Facial Recognition – Why Is It A Matter Of Concern For Several States And City Governments?

Info 850x500 2

Source: Facial Recognition: Changing the Face of Travel by Katherine Noyes – WSJ

Facial detection, and if further enhanced, facial recognition is growing into a staple technology of the near future for many departments. The potential it carries, however, is not convincing enough for those who are concerned with its risks. For example, as per a survey by Reservations.com, over 32.5% of US citizens do not approve of the use of this technology at airports for better security and faster boarding, due to the privacy concerns that it brings with it.

The breach of privacy of individuals, the presence of individuals such as minors and those not concerned with a matter at hand, such as bystanders in a crime scene, as well as the exposure of PII (Personally Identifiable Information) information, all are matters which have led several US states, cities, and law enforcement jurisdictions to rethink the implementation of facial recognition technology.

This becomes a greater issue when video footage is being presented on a public medium, played as evidence in court, or accessed and viewed in some other manner, rather than when at rest in storage.

Automatic Face Redaction – How Can We Counteract These Issues?

One of the solutions to counteract the issues we discussed is the implementation of redaction. Face redaction would allow us to obfuscate the faces of individuals present in the video footage at hand. To do this, however, initially, a face detection computer vision algorithm would have to be executed to detect all the faces present in the video, which would later be redacted.

An example of Face Redaction

How Does Automatic Redaction Of Faces Help Broaden The Implementation Of Face Recognition?

With the implementation of automatic face redaction, all faces present in the video will be redacted automatically. This helps us meet the general compliance requirements set by law enforcement and regularity firms, which redaction serves to fulfill.

The CJIS, CRoC, HIPAA, FEDRamp, CCPA, GDPR and the California State Bill 748 (SB-748) are amongst those that emphasize redaction for all video evidence and media.

How Does Automatic Face Redaction Work?

Face redaction initially works with the placement of a layer over the video stream, where the masking is done. Using this layer, selections are made where faces appear onscreen. These selections, or bounding boxes, are then filled with an opaque color, blurred out or pixelated, to obfuscate the face within. This is how manual face redaction works.

If the process is automated, an AI algorithm allows for facial detection, which can later be redacted with ease with a few clicks. Facial recognition, as mentioned earlier, adds further to the process by making it more efficient in cherry-picking whom to redact and who to not redact.

VIDIZMO’s Redaction Tool –The Cutting-Edge Solution For Automatic Redaction

Countless tools are available for use in the market for redaction. However, only a few of the best redaction tools provide automatic redaction for video, audio and imagery. Each of these has varying parameters that make them compete against one another.

VIDIZMO’s Redaction Tool is amongst these names of one of the leading redaction tools in the market, in terms of services offered. That is because VIDIZMO’s redaction tool provides automatic video and audio redaction through on-demand face detection, making it easier for you to redact videos when ingesting several terabytes of data from multiple sources.

Apart from that, the tool also holds a multitude of features that make it one of the best the market has to offer.

  • VIDIZMO’s Redaction Tool is available as a standalone tool for the person intending to only redact videos. For large enterprises and law enforcement agencies, it can be opted for as an add-on to the complete solutions, the EnterpriseTube Video Content Management System (EVCMS) and the Digital Evidence Management System (DEMS).
  • VIDIZMO utilizes the power of AI for all its services and features, including redaction. The results provided by VIDIZMO’s AI are efficient, accurate and on-par with market standards, if not better.
  • The tool is efficient in automatically redacting faces and people, and can also be used to redact PII information, vehicles and license plates, objects and more.
  • The tool is suitable for audio redaction as well, where it will mute the redacted segment of an audio or override it with a bleep sound.
  • VIDIZMO ensures the security of your data through the provision of a complete system for video content management.
  • Easy-to-use, the software developed by VIDIZMO are all browser-based, and packed with a simple user interface, making it possible for every user to have no difficulty in using the tool.
  • With active tracking of detected elements, VIDIZMO’s redaction tool saves you from the hassle of masking elements for individual frames.
  • You can either blur, pixelate or hide an object with a solid, opaque box to redact it.
  • The objects detected can be renamed to a more specific name, which acts as metadata. This allows the file to show up in relevant, enhanced searches and more!
  • With multiple deployment options to cater to the needs of various customers, VIDIZMO is available as SaaS, government cloud, commercial cloud or in an on-premises datacenter of your preference.
  • The tool gives the user the liberty to choose between retaining the original file or overwriting it with the redacted file.
  • The redaction tool by VIDIZMO provides integration options for your preexisting systems such as an RMS or CMS, through APIs and widgets.
  • VIDIZMO’s Redaction tool is FedRAMP compliant through our partnerships.
  • The tool allows you to efficiently process and redact multiple files simultaneously.

If you wish to learn more about our online redaction tool, be sure to check out the video redaction tool page or contact us.

Contact Us

To learn more on the topic, do check out our detailed guide on redaction.

Posted by Muhammad Nabeel Ali

Nabeel is a technology content strategist at VIDIZMO and an expert in evidence management technologies. He is actively researching on innovative trends in this domain such as AI. For any queries, feel free to reach out to nabeel.ali@vidizmo.com

VIDIZMO Whitepapers

Post a comment