Sumsub is a full-cycle verification platform that secures every step of the user journey. With Sumsub’s customizable KYC, KYB, AML, Transaction Monitoring and Fraud Prevention solutions, you can orchestrate your verification process, welcome more customers worldwide, meet compliance requirements, reduce costs and protect your business.
Sumsub achieves the highest conversion rates in the industry—91.64% in the US, 95.86% in the UK, and 97.89% in Hong Kong—while verifying users in less than 50 seconds on average.
Sumsub’s methodology follows FATF recommendations, the international standard for AML/CTF rules and local regulatory requirements (FINMA, FCA, CySEC, MAS, BaFin).
Sumsub has over 2,000 clients across the fintech, crypto, transportation, trading, e-commerce and gaming industries including Bitpanda, Wirex, Avis, Bybit, Huobi, Kaizen Gaming, and TransferGo.
Learn more
PackageX OCR API converts any smartphone into a powerful universal label scanner that reads every bit of text on the label, including barcodes and QR codes.
Our state-of-the-art OCR technology uses robust deep learning models and proprietary algorithms to extract information from package labels.
Our OCR API is trained based on information from over 10 million labels, enabling over 95% scan accuracy -- the best in the market.
Our technology scans in low-light conditions, reads at any angle, and works with damaged labels.
Build your custom OCR scanner app and remove pen-and-paper inefficiencies.
Easily extract information from both printed text and handwritten labels with our OCR scanner.
Our OCR technology is trained on multilingual label data extracted from over 40 countries.
Detect & extract information from any barcode or QR code.
Learn more
VeriLook SDK
VeriLook SDK delivers high-performance facial identification for stand-alone or client-server biometric systems. Built for security, access control and identity verification, it performs fast and accurate matching in 1-to-1 and 1-to-many modes. Its ISO 30103-7 compliant liveness detection prevents spoofing by distinguishing real faces from printed images.
The SDK recognizes multiple faces at once, handles partially covered faces and extracts attributes including age, gender, facial hair, glasses and expressions. It works with images captured in visible or near-infrared light and maintains reliable accuracy across lighting changes and varied head positions.
Developers can integrate VeriLook using C/C++, C#, VB .NET, Java or Python on Windows, Linux, macOS, iOS and Android.
With more than twenty years of deployments worldwide, VeriLook offers a dependable, flexible and developer-friendly platform for building secure, modern biometric applications.
Learn more
