Building Advanced Face Search Workflows with Didit's Web SDK

Seamless IntegrationDidit's Web SDK allows for straightforward integration of advanced Face Search functionalities, enabling developers to quickly build robust identity verification workflows.

Powerful Fraud PreventionLeverage 1:N Face Search to automatically detect duplicate accounts and cross-reference against blocklists, significantly enhancing your fraud prevention strategies.

Configurable SecurityCustomize Face Search sensitivity with configurable similarity thresholds, empowering you to balance user experience with your organization's specific risk tolerance.

Didit's AdvantageDidit provides an AI-native, modular platform with a free core KYC, making advanced biometric features like Face Search accessible and scalable for businesses of all sizes, with no setup fees.

The Power of 1:N Face Search in Modern Applications

In today's digital landscape, verifying user identity is paramount. Beyond simply confirming an individual's identity, businesses face the challenge of detecting sophisticated fraud attempts, preventing duplicate accounts, and ensuring the overall integrity of their user base. This is where 1:N Face Search becomes an invaluable tool. Unlike 1:1 face matching, which compares a face against a single reference, 1:N (one-to-many) Face Search scans a submitted face against an entire database of previously verified identities. This capability is crucial for identifying individuals attempting to create multiple accounts, or those who have been previously blocklisted.

Didit's 1:1 Face Match & Face Search capabilities are designed to provide high accuracy and rapid results, even with large user databases. By integrating these features into your applications, you can proactively combat fraud, streamline user management, and build a more secure digital environment. The ability to automatically check against blocklists further strengthens your defenses, ensuring that known problematic users cannot easily re-enter your system.

Integrating Face Search with Didit's Web SDK

Didit's Web SDK makes it incredibly easy for developers to embed sophisticated Face Search functionality directly into their web applications. The SDK provides a clean, API-driven approach, allowing for seamless integration without needing deep expertise in biometric algorithms. The process typically involves capturing a user's facial image via the SDK, which then securely transmits it to Didit's AI-native platform for analysis.

Upon submission, Didit's system extracts unique facial features and compares them against your existing database of verified faces. The results, including similarity scores and potential matches, are returned via a structured JSON report. This report not only identifies potential matches but also provides crucial details such as session IDs, verification dates, and user information, all while adhering to strict security protocols like temporary image URLs for matched faces.

For instance, a developer building an online gaming platform could use Didit's Web SDK to capture a new user's selfie during registration. This image would then be automatically subjected to a 1:N Face Search against all existing player profiles. If a high similarity match is found, indicating a potential duplicate account, the system can flag it for review or automatically decline the registration, preventing bonus abuse or multi-accounting. The modular architecture of Didit allows for this type of integration to be highly customizable to fit specific use cases.

Configuring and Understanding Face Search Results

One of the key advantages of Didit's Face Search is its configurability. Businesses can customize search sensitivity by setting specific similarity thresholds. A higher threshold, for example, 90%, will yield fewer but more precise matches, suitable for high-security applications where false positives must be minimized. A lower threshold, say 70%, will cast a wider net, potentially identifying more matches but with a higher chance of false positives, which might be acceptable for initial screening processes.

The Face Search report provides comprehensive details for each match, including a similarity_percentage, session_id of the matched user, and critical user_details. Importantly, the report also indicates whether a matched face is is_blocklisted. This direct integration with Didit's blocklist feature means that if a user's face matches an entry in your blocklist, the verification can be automatically declined, providing an immediate layer of defense against known fraudsters.

Developers should also be aware of potential warnings that can arise during the face search process, such as NO_FACE_DETECTED or MULTIPLE_FACES_DETECTED. Didit provides clear warning tags and descriptions, allowing developers to build robust error handling and user guidance into their workflows. For cases where multiple faces are detected, applications can be configured to either fail the session or return the largest face in the image for processing, offering flexibility based on specific requirements.

Advanced Use Cases and Fraud Prevention

Beyond simple duplicate account detection, Didit's Face Search opens up advanced fraud prevention strategies. Consider an e-commerce platform that frequently deals with returns or chargebacks. By performing a Face Search during account creation, they can identify individuals who have previously engaged in fraudulent behavior under different email addresses or names. This proactive approach saves significant resources and protects revenue.

Another powerful application is in compliance and regulatory environments. For instance, financial institutions can use Face Search to ensure that individuals are not attempting to open multiple accounts to circumvent AML (Anti-Money Laundering) regulations. When combined with Didit's AML Screening & Monitoring, this creates a comprehensive shield against financial crime.

The API access to Face Search functionality means that businesses are not limited to pre-defined workflows. They can programmatically submit face searches, integrate matching capabilities into custom fraud detection engines, and build automated systems that adapt to evolving threats. This developer-first approach, coupled with Didit's AI-native capabilities, ensures that your identity verification infrastructure is always at the cutting edge.

How Didit Helps

Didit stands out as the premier solution for building advanced Face Search workflows. Our AI-native platform offers unparalleled accuracy and speed in 1:N Face Search, enabling businesses to effectively detect duplicate accounts and prevent fraud across their user base. With Didit's modular architecture, you can easily integrate Face Search as a plug-and-play identity check into your existing systems using our clean APIs or no-code Business Console.

Didit's commitment to a developer-first experience means instant sandboxes and comprehensive public documentation, making integration straightforward. We provide configurable thresholds for Face Search, allowing you to fine-tune the sensitivity to match your specific risk tolerance. Furthermore, our seamless integration with the blocklist feature ensures that previously identified problematic users are automatically flagged and prevented from creating new accounts. All of this is available with Didit's free tier for core KYC, emphasizing our dedication to making robust identity verification accessible, with no setup fees.

Ready to Get Started?

Ready to see Didit in action? Get a free demo today.

Start verifying identities for free with Didit's free tier.