#AWS Face Compare Pro
**Documentation last updated:** March 2026 | **Plugin last updated:** March 2026

# Overview

Instantly and securely compare two faces using Amazon Web Services (AWS) Rekognition inside your Bubble.io application.

This plugin allows you to build robust KYC (Know Your Customer) flows, automate identity verification, and prevent fraud by comparing a user's live selfie against a government-issued ID card or passport photo.

Unlike other plugins, this runs directly as a Server-Side Action (Backend), meaning your API keys are never exposed to the client, and you can process images asynchronously using standard HTTPS URLs or private AWS S3 bucket URIs.

---

# Pre-Requisites & Requirements

Before you can use this plugin in your Bubble application, you MUST have an active Amazon Web Services (AWS) account.

Because this plugin executes entirely on the backend for maximum security, you do NOT need a complex Cognito Identity Pool as you do for front-end streaming cameras. You only need a simple IAM User with programmatic access.

---

# How to use (Step-by-Step Setup)

Follow these exact steps to configure your AWS environment and integrate the plugin into your Bubble app.

### Step 1: Create AWS IAM Credentials (For the Backend Action)
Your Bubble server needs permanent, secret keys to securely send images to AWS for algorithmic comparison.
1. Log in to your AWS Console and search for **IAM** (Identity and Access Management).
2. Navigate to **Users** -> **Create user**. Name the user `BubbleFaceComparisonBackend`. Do *not* grant AWS Management Console access.
3. On the Permissions page, select **Attach policies directly**.
4. Click **Create policy**. Switch to the **JSON** editor and paste this exact security policy:
```json
{
   "Version": "2012-10-17",
   "Statement": [{
       "Effect": "Allow",
       "Action": [
           "rekognition:CompareFaces"
       ],
       "Resource": "*"
   }]
}
```
*(Optional: If you intend to pass private `s3://` URIs instead of public HTTP links to the plugin, you must also add `"s3:GetObject"` permissions to this policy for your specific bucket).*

5. Name the policy `BubbleFaceComparisonPolicy` and save it. Return to the User creation tab and attach this newly created policy.
6. Once the user is created, click on the user's name, go to the **Security credentials** tab, and click **Create access key**.
7. Choose "Application running outside AWS".
8. Copy the **Access Key ID** and the **Secret Access Key**. Store them safely.

### Step 2: Configure the Plugin in Bubble
1. Open your Bubble.io application editor.
2. Navigate to the **Plugins** tab and select the AWS Face Comparison Pro plugin.
3. In the plugin settings, paste your **AWS Access Key**, **AWS Secret Key**, and your default **AWS Region** (e.g., `us-east-1`).

---

# Components (Actions)

*(Note: Because this is a utility plugin that runs algorithms on images you provide, there are no visual front-end Elements to drag onto the page. It operates entirely as a Workflow Action).*

### Action #1: AWS Compare Faces (Back-End)
*Use this action in a workflow anytime you need to analyze two images.*
*   **Description:** Securely transmits two images (a Source and a Target) to AWS Rekognition to determine if the largest face in the Source image matches any faces in the Target image.
*   **Input Fields:**
   *   **Source Image URL (Text):** The baseline image (e.g., an ID Card or established profile picture). Must be a valid `https://` Protocol URL, a Bubble uploader link (e.g., `//s3.amazonaws.com/...`), or a private AWS `s3://bucket/key.jpg` URI.
   *   **Target Image URL (Text):** The image to check against the baseline (e.g., a new Selfie).
   *   **Similarity Threshold (Number 0-100):** Specifies the minimum strictness level for a match. AWS defaults this to 80 if left blank. If you want high security, enter 90 or 95. If the faces match below this threshold, it will register as a failed match.
   *   **AWS Region Override (Text):** Overrides the global plugin region setting if you need to route requests elsewhere temporarily.
*   **Returned Values:**
   *   **Is Match (Yes/No):** A definitive boolean determination. "Yes" means AWS is confident it is the same person based on your threshold.
   *   **Similarity Score (Number 0-100):** The exact percentage of confidence that the faces belong to the same person (e.g., 99.98).
   *   **Source Faces Detected (Number):** How many faces were found in the first image. Useful for error checking (e.g., if this is 0, the user uploaded a picture of a wall instead of an ID).
   *   **Target Faces Detected (Number):** How many faces were found in the second image.

---

# Workflow Integration Examples

How to orchestrate a KYC (Know Your Customer) approval flow using standard Bubble elements and this plugin.

### Step A: The Setup
1. Place two `Picture Uploader` elements on your Bubble page. Name one "ID Uploader" and the other "Selfie Uploader".
2. Create a "Submit for Verification" Button.

### Step B: The Comparison Workflow
Trigger this when the user clicks the "Submit for Verification" button.
1.  **Plugin Action:** Run `AWS Compare Faces`.
   *   Set `Source Image URL` to `ID Uploader's value's URL`.
   *   Set `Target Image URL` to `Selfie Uploader's value's URL`.
   *   Set `Similarity Threshold` to `90`.
2.  **Conditionals (Success Path):** Add a subsequent Action (e.g., "Make changes to Current User" -> `Is Verified = Yes`) *Only When* `Result of Step 1's Is Match` is "yes".
3.  **Conditionals (Failure Path):** Add a subsequent Action (e.g., "Show an Alert -> Face Mismatch") *Only When* `Result of Step 1's Is Match` is "no".
4.  **Error Handling (No Faces):** Add an Action to alert the user if they uploaded a bad photo *Only When* `Result of Step 1's Source Faces Detected` is `0`.

---

# Plugin Demo

[Bubble plugins](https://bubble.io/plugins)

**[View the Live Interactive Demo Here](https://aws-plugin-preview.bubbleapps.io/version-test/face_comparision)**

**[View the Demo App Editor Here](https://bubble.io/page?id=aws-plugin-preview&tab=Design&name=face_comparision&type=page)**

---

# Contact / Support

For feature requests, bug reports, or if you need assistance configuring your AWS IAM policies for private S3 Bucket access, please reach out directly:

📧 **[[email protected]](mailto:[email protected])**