The rapid evolution of digital services and the increasing global connectivity has come along with a challenge of ensuring authentic identity. Paravision is building Vision AI tools to support establishing trusted identity for customers, employees, or citizens in remote and digital environments.
With the relative ease of creating physical reproductions or digital manipulations, matching one face to another with highly accurate face recognition is not enough to prove that a presented identity is authentic.
Authentic Identity is a collection of technologies, systems, policies, and processes to create trust around identity in both physical and digital domains.
Challenges to Authenticity, Identity, and Trust
Physical Presentation Attacks
Also known as Spoofs, Physical Presentation Attacks can include paper cutouts, digital displays, latex or 3D-printed masks used with the intention of identity fraud
Digital Face Manipulations
Digital Face Manipulations can include deepfakes that use AI to replace the likeness or expression of one person with another in digital media
Presentation attacks in a digital domain with synthesized, hyper-realistic faces that do not exist in the real world, such as those created by StyleGAN
3D NIR Liveness
Advanced liveness for embedded devices
Our most advanced Presentation Attack Detection for Ambarella CVFlow uses a combination of near-infrared and other sensing to deliver real-time, edge-based PAD for a wide range of 2D and 3D spoof attacks.
2D/RGB Passive Liveness
Intuitive PAD for standard cameras
Deployable on mobile devices or server-side, Paravision’s 2D/RGB Liveness (PAD) is able to process single images and deliver guidance on presentation attacks in use cases where identity authentication relies on 2D RDB image, such as you would get from a smartphone selfie.
Tackling an emerging threat
From identity swaps and expression swaps to synthetic faces, face morphs, and adversarial attacks, digital presentation attacks are broad in nature and pose a serious risk to security, safety, and privacy. Paravision is proud to be building tools to detect them.
Paravision Presentation Attack Detection Product Suite
Ensuring the authenticity of biometric images is critical to high confidence biometric identification and authentication in unattended or fully automated use cases, from digital ID verification to phy...
Paravision Face Recognition and Liveness SDKs for Mobile Platforms
From consumer-facing IDV and eKYC applications to ruggedized tablets for government identity programs, from next-generation kiosks for point-of-sale and air travel to next generation access control an...
Authentic Identity: Challenges and Opportunities in Physical and Digital Domains
Presented identities can no longer be expected to be authentic identities. This creates risks for democracy, national security, business, human rights, and personal privacy. In this paper, we will exp...