Paravision Launches Age Estimation Tool

Paravision has officially launched its own age estimation tool based on face biometrics. The solution is available in the form of a Software Development Kit or packaged in Docker-based products, and can be used independent of Paravision’s face-matching technology, helping to ensure end users’ privacy.

The product launch comes after Paravision’s announcement last fall that it had teamed up with Persona, a San Francisco-based identity verification startup, on a planned “age estimation and verification” tool. At the time, the companies explained that their joint solution would estimate an end user’s age based on a selfie image, and that it would use Persona’s “granular data retention and redaction policies” to protect privacy.

The announcement of “Paravision Age Estimation” made no reference to Persona, but did maintain an emphasis on “a privacy-centric approach”. Paravision explained that its solution could prove useful in social media, gaming platforms, and other online communities, especially given an evolving regulatory landscape that includes the Kids Online Safety Act in the US and the Online Safety Act in the United Kingdom.

True to form, Paravision also emphasized its commitment to ethical AI development, which has been a key focus for the company for some time.

“I’m incredibly proud of the technical rigor and ethical commitment of our Age Estimation team,” said Paravision’s Chief Technology Officer, Charlie Rice. “Our solution is ethically trained on a proprietary dataset of hundreds of thousands of properly-consented images, ensuring broad race, gender, and diversity. It is designed to keep people safe while preserving their privacy.”

Paravision’s launch of its own age assurance solution speaks not only to a growing market demand for such technologies, but also to Paravision’s commitment to expanding its own technological portfolio. Earlier this year, the company launched a new liveness detection system tailored for selfie-based identity verification, as well as a deepfake detection tool to combat the growing threat of synthetic face images.

–

March 19, 2024 – by Alex Perala