Overview
VIDA offers multiple liveness detection modes to ensure robust and secure user verification across various use cases. Supported modes include:
Each mode is designed to counter different types of spoofing attempts and can be configured based on your application’s security requirements.
Passive Liveness
Passive Liveness operates in the background, analyzing selfies with advanced AI and ML techniques to confirm the image's authenticity without requiring user interaction. It generates a selfie score from 0 to 1 , basis on which the genuineness of the selfie is determined. By default, passive liveness is enabled.
Gesture Liveness
Gesture Liveness involves user interaction, where the system prompts the user to perform simple actions like blinking, smiling, or moving their head. This method ensures that the user is live and present, making it harder for attackers to spoof the system.
Requires a simple user action (such as a head movement). This method includes checks for injection attacks, image quality, liveness, and image manipulation.
Zoom Liveness
Zoom Liveness requires a series of gestures including eyeblink, smile, head shake, zoom gesture, and face frame validation. These actions provide stronger security by enabling deeper analysis of motion, depth, and facial cues.
Color Flash
Color Flash-based liveness works by flashing specific colors on the screen and analyzing the reflection of those colors on the user's face in real time. This ensures that the user is physically present and not a static image or replay attack, enhancing the robustness of the liveness check through light-based interaction.
Comparison between All Liveness
Checks carried out
Image Quality check
Liveness check and Image Manipulation Check
All in passive liveness, and:
Eye Blink Check
Smile Check
Head Movement Check
All in passive liveness, and:
Face Zoom Check
Eye Blink Check
Smile Check
Head Movement Check
All in passive liveness, and:
Reflection Check
Color Check
Liveness HD
Benefit
Easy to implement and convenient for users
The photo taking process is faster
Provides higher security because it requires active user interactions (physical movements or actions).
Hard to imitate using masks, printed photos, screen replays, or AI-generated selfies.
Difficult to spoof with pre-recorded videos since the prompts are random and unpredictable.
Adds motion-based depth analysis for spoof resistance
Simple and intuitive for users to follow
Effective against printed photos and screen replays
Simplifies liveness verification by removing the need for active gestures (blinking, smiling, nodding).
Users only need to look at the screen while a sequence of colors flashes.
Enhances security by detecting depth, skin texture, and dynamic light response.
Image Manipulation Detection
To further enhance security, VIDA incorporates an Image Manipulation Detection check as an additional layer of scrutiny. Learn more about this process on our Image Manipulation Detection.
Liveness Threshold
VIDA also applies a default liveness threshold of 0.95, which determines whether a selfie is accepted as live or rejected as spoofed based on the liveness score range of 0.0–1.0.
Image Quality Assessment
VIDA evaluates the quality of captured images to minimize false rejections due to poor image quality. The VIDA image quality standard adheres to ISO/IEC 29794-5.
Under exposure
Error response feedback
Real time feedback
Over exposure
Error response feedback
Real time feedback
Blur
Error response feedback
Real time feedback
Eye visibility
Error response feedback
Real time feedback
Occlusion prevention
Error response feedback
Real time feedback
Eyes open
Error response feedback
Real time feedback
Crop of face image prevention
Error response feedback
Real time feedback
Last updated