> For the complete documentation index, see [llms.txt](https://docs.vida.id/identity-stack/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.vida.id/identity-stack/verify/kyc-mobile-sdk/sdk-user-journey/selfie-capture.md).

# Selfie Capture

## Face Auto-Capture

In this step, the user takes a selfie, which is then used for both liveness detection and biometric face comparison. The SDK ensures that the captured selfie meets quality standards by guiding the user with real-time prompts and automatically capturing the best possible frame. This simple approach makes the process user-friendly while still producing a reliable input for downstream verification.

* Automatic detection and capture of the best-quality face frame.
* Guidance to reduce movement which can avoid blur, glare, or off-angle shots.

## Liveness Detection

The captured selfie is analyzed to confirm that it comes from a live person rather than a static image, video, or mask. Depending on the configuration, the SDK can apply passive checks, ask for simple gestures like blinking or turning the head, or use color flash techniques. By combining these methods, the SDK helps ensure that the user is genuinely present during the capture, strengthening security against spoofing attempts.\
\
Multiple methods supported:

* **Passive Liveness:** Camera-based signals with no user action.
* **Gesture Liveness:** User prompts (blink, head turn).
* **Color Flash Liveness:** Dynamic light patterns validated by device camera.

## Biometric face comparison

Once liveness is confirmed, the selfie is compared against a reference image to ensure identity consistency. Clients can configure the SDK to match the selfie with the **portrait extracted from the identity document** or, in some cases, with an **authoritative database** such as the Indonesian identity system. This flexibility allows organizations to tailor the verification process to their security and compliance requirements.

* Selfie ↔ Document portrait.
* Selfie ↔ Authoritative source (support for Indonesian national identity database).

## Face Blacklist Search 

Please refer to this page [here](https://docs.vida.id/vida-identity-platform/~/revisions/TOFdPL7RFt4edkIllkrk/integration-methods/api/biometric-verification/api-reference/blacklist-enrollment-and-search)

To further strengthen fraud detection, the SDK can also perform a **deduplication check**. Here, the user’s selfie is compared against a larger database of previously enrolled faces. If a duplicate is found, the system flags the potential fraud attempt, preventing the same person from registering multiple times with different documents. This optional 1:N check gives clients an added layer of assurance in high-risk environments.

* Backend duplicate search to detect repeated enrollment attempts.
* Prevents identity recycling across multiple accounts.