Online Video KYC Reimagined by Team Heisenberg for Standard Chartered Hackathon 🦅

Team_Heiseneberg_Presentation.mp4

Problem Statement 🛠️

Currently, Know Your Customer (KYC) processes are predominantly offline or non-interactive online processes. These methods are often manual, cumbersome, and lack user engagement. They can be particularly challenging for users with limited financial resources or education levels. Additionally, existing processes may not be inclusive across languages and generations.

We aim to address these challenges by creating an online interactive Video KYC process. This process should be intuitive, self-sufficient, and highly inclusive. It should capture the user's live photograph and basic details such as name, date of birth (DOB), address, PAN card/Aadhaar, signature, income range, and type of employment in a conversational manner. Our goal is to bridge the gap across languages and generations, ensuring accessibility for all users regardless of their financial status or educational background.

Solution 💪🏽

Our platform aims to streamline and automate the Video KYC process, making it faster, more user-friendly, and compliant. We propose the following features:

Current Standard Charted Video KYC Process 👩🏻‍💻

• Email or SMS sent with Video KYC Link
• Permissions Check (Location Access, Audio/Video Capability, Internet Connectivity, Internet Speed)
• Language Selection and OTP Generation for consent
• Video/Audio Recording
• Facial Authentication
• PAN Photo Capture
• Appointment Scheduling
• Browser/tab locking during KYC Process

Our Solution 📝

• AI-driven conversational flow
• Facial Authentication
• PAN Photo Verification
• Signature Verification
• Agent Support throughout KYC Process
• Read-aloud Details with Voice Confirmation
• Multiple Language Support
• ChatBot Support

Architecture 🛠️

Future Scope 📈

• Implement a step-by-step follow model for senior citizens
• Continuous improvement of AI algorithms for better accuracy and user experience
• Integration with additional identity verification methods (e.g., biometrics)

Team Members

• Aryan Raj (ML and Backend Developer)
• Divyanshu Yadav (Full Stack Developer)
• Puranjay Bhargava(ML and Backend Developer)
• Kevin Paul Jacob (Full Stack Developer)
• Mathangy K (ML Developer)
• Mohammad Sami (Full Stack Developer)

Tech Stack 💻

Frontend 🥷🏼

• Next.js
• WebRTC for real-time video communication
• Shadcn
• Crisp Customer Support

Backend 🐐

• Python with Flask for RESTful API development
• Express.js
• NOde.js
• MongoDB
• TypeScript
• MDN Web API

Machine Learning 🧠

• VGG 19 for facial recognition
• Opencv
• pytesseract
• CVQRScanner
• Deepface
• cascade classifier
• Bark and whisper AI

Conclusion

Our Video KYC Automation Platform aims to revolutionize the KYC process, making it seamless, secure, and compliant while providing a superior user experience. We welcome contributions and feedback from the community to further enhance and improve our platform.