Making Video Anonymous by Default


This post is also available in: עברית (Hebrew)

The growth in the capture of video in public places from CCTV, body-worn police cameras and infrastructure surveying (roads, streets, buildings) has created a critical need for accurate and fast redaction services to protect the personal data that is being captured, such as faces, number plates, house numbers, body markings etc.

Organizations hold thousands of hours of video. With facial recognition technologies, a video showing you were in a certain place at a certain time reflects a biometric and unchangeable aspect of identity and reveals recognizable personal data. But it’s also one of the hardest to anonymize manually, requiring big teams to trawl through footage, and with a significant risk of missing something. Facial recognition is nothing new – but in most cases, the algorithms are trained to identify and profile individuals, not to remove them.

Based on machine-learning, a new platform makes video anonymous by default. SecureRedact ensures no frames of facial privacy are missed. The technology provides fast and accurate redaction of security, survey and events footage in compliance with GDPR (European Union privacy regulation) and Data Privacy legislation, according to

The platform enables various sectors, from law enforcement agencies to security, smart cities, and global brands, to use machine-learning to make their video anonymous in an instant.

The innovative technology developed by UK-based Pimloc is available in the cloud, with the option to host on-premise where needed. The platform has been developed to thrive against even the most messy and difficult to process video, from sources like CCTV and vehicle-mounted cameras in all weather and lighting conditions.

The AI platform which has been trained on millions of images and video has been further fine-tuned with domain-specific video from security and road survey footage. It can be refined by learning from each organization’s specific data to improve this further.