# What is Selfie Intelligence?

Selfie Intelligence provides a selfie-only verification flow using Socure’s Capture App or SDKs. It enables organizations to collect a live selfie from a user without requiring any ID documents and applies advanced biometric and fraud-detection checks to determine whether the selfie is genuine and captured in real time.

Selfie Intelligence is designed for fast, low-friction verification scenarios—such as age checks, security validation, and trust & safety workflows—where a live selfie provides the assurance needed without the overhead of document capture.

# How it works

- **Live selfie capture:** The user captures a selfie using the Capture App or SDKs.
- **Liveness detection:** The system verifies that the selfie was taken by a live person, not a spoof (photo, screen, video, mask, etc.).
- **Injection attack protection:** Detects attempts to bypass the camera by feeding manipulated or pre-recorded images into the capture flow.
- **Selfie age estimation:** Produces an estimated age, enabling age-gating or policy checks.
- **Digital Intelligence:** Device and behavioral signals help identify risky or atypical activity.
- **Global coverage**: Available in 180+ countries worldwide wherever selfie capture workflows are supported.

# Verification checks

Selfie Intelligence applies the following checks to each captured selfie:

| Check                        | Description                                                                                         |
|------------------------------|-----------------------------------------------------------------------------------------------------|
| Liveness detection            | Confirms the selfie is captured by a live person, not a photo, video, screen, or mask.             |
| Injection attack detection     | Identifies attempts to bypass the camera by feeding pre-recorded or manipulated images into the capture flow. |
| Selfie age estimation         | Returns a `predictedAge` value based on the captured selfie image, enabling age-gating and policy checks. |
| Digital Intelligence          | Evaluates device integrity and behavioral signals to detect atypical or risky activity during the selfie session. |
| Decision + reason codes       | Returns a structured decision (`accept`, `review`, or `reject`) with reason codes for workflow routing. |

# Consumer flow

The following diagram shows the screens a consumer sees during the Selfie Intelligence flow in the Capture App:

```
Retake

Submit

Splash Screen

Terms & Consent

Selfie Capture

Selfie Preview

Confirmation Screen
```

1. **Splash Screen** — The first screen the consumer sees when redirected to the Capture App.
2. **Terms & Consent** — A mandatory notice and consent screen. This screen cannot be removed, customized, or disabled.
3. **Selfie Capture** — The consumer captures a live selfie using the device's front-facing camera. If auto-capture is enabled, the system automatically takes the selfie when the face is positioned correctly.
4. **Selfie Preview** — The consumer reviews the captured selfie and confirms it's clear, or retakes the photo.
5. **Confirmation Screen** — Confirms the selfie was submitted successfully.

For screenshots of each screen, see [Capture App flow examples](https://help.socure.com/riskos/docs/example-capture-app-flows#/).

# What you can do with Selfie Intelligence

- **Enable selfie-only onboarding** without requiring document uploads.
- **Verify age** for age-restricted products, services, or features.
- **Add liveness checks** to ensure real-person interaction during sensitive actions.
- **Strengthen trust & safety** by detecting spoofing and manipulated media.
- **Support low-friction security workflows** where a live selfie is sufficient for authentication.

# Unique features

- **Selfie-only verification:** No document capture required.
- **Liveness analysis:** Ensures the selfie is captured from a live person.
- **Injection attack protection:** Identifies attempts to bypass the camera or upload tampered images.
- **Selfie age estimation:** Returns an estimated age for age-restricted scenarios.
- **Digital Intelligence:** Device and behavioral signals provide additional fraud insights.
- **Decision + reason codes:** Clear outputs for routing in RiskOS™ workflows.

# Integration options

Integrate with the Predictive Document Verification SDKs to embed the [**Capture App**](https://help.socure.com/riskos/docs/capture-app#/) into your mobile application, then verify a consumer’s identity by calling the Evaluation endpoint.

The **Capture App** provides a guided, mobile-optimized workflow for capturing IDs, selfies, and supporting documents.

[**iOS SDK**](https://help.socure.com/riskos/docs/ios-sdk#/)

[**Android SDK**](https://help.socure.com/riskos/docs/android-sdk#/)

[**React Native**](https://help.socure.com/riskos/docs/react-sdk#/)

[**Web SDK**](https://help.socure.com/riskos/docs/web-sdk#/)

# Example response

The following is an example of the `selfieIntelligence` object returned from the [Evaluation endpoint](https://help.socure.com/riskos/docs/docv-sdk-integration#/):

```json
{
  "selfieIntelligence": {
    "decision": {
      "strategy": "standard",
      "value": "accept"
    },
    "reasonCodes": [
      "I834"
    ],
    "digitalIntelligence": {
      "decision": {
        "strategy": "standard",
        "value": "accept"
      },
      "reasonCodes": [
        "I401"
      ]
    },
    "signals": {
      "selfie": {
        "predictedAge": 23
      }
    }
  }
}
```

| Field                       | Description                                                                                                               |
|-----------------------------|---------------------------------------------------------------------------------------------------------------------------|
| `decision`                  | The overall Selfie Intelligence decision (`accept`, `review`, or `reject`).                                              |
| `decision.strategy`         | The configured decision strategy (for example, `"standard"`).                                                          |
| `reasonCodes`               | Reason codes explaining the decision. See the [reason code list](https://help.socure.com/riskos/docs/reason-codes) for descriptions. |
| `digitalIntelligence`       | Device risk signals captured during the selfie flow, including a separate decision and reason codes.                     |
| `signals.selfie.predictedAge` | Estimated age of the individual based on the captured selfie.                                                            |

# Next steps

[RiskOS™ Setup](https://help.socure.com/riskos/docs/selfie-intelligence-setup#/)

[Capture App Setup](https://help.socure.com/riskos/docs/customization-branding-copy#/)

[API Integration](https://help.socure.com/riskos/docs/docv-sdk-integration#/)

Updated about 5 hours ago