TimeCapsule AI: Reviving Historical Figures Through AI

Project Domain / Category

Artificial Intelligence (AI) / Deep Learning / Image & Video Processing

Abstract / Introduction

This project aims to bring historical figures to life using AI-driven facial animation and speech synthesis. The system will take old photographs or paintings of historical figures and animate them, making them appear as if they are speaking. AI will generate realistic facial movements and synced speech, allowing users to experience interactive storytelling in education, museums, and documentaries.

The system will allow users to upload an image of a historical figure, input text, and generate a video where the person appears to be narrating their own story. This project will help in education, museums, documentaries, and digital preservation by making learning more engaging and interactive.

Historical Figures Animated Using AI Technology Example: https://www.youtube.com/watch?v=YPjLycUANZs

Functional Requirements:

1.      The system should allow users to upload an image of a historical figure.

2.      The system should animate the face by adding realistic facial movements and expressions.

3.      The system should synchronize lip movements with AI-generated speech.

4.      Users should be able to input text, and the system will generate a video where the historical figure speaks.

5.      The system should provide an exportable video file for users.

6.      Provide a pre-set library of historical figures for demonstration (optional).

Tools:

·        Programming Language: Python

·        AI Models: First Order Motion Model (for animation), Wav2Lip (for lip syncing), ElevenLabs (for AI voice synthesis)

·        Libraries & Frameworks: OpenCV, TensorFlow/PyTorch, DeepFaceLab

·        Development Environment: Jupyter Notebook, Google Colab, or VS Code

·        Deployment (Optional): Flask/Django for a web-based interface

Supervisor:

Name: Sonia Salman

Email ID: sonia.salman@vu.edu.pk

Skype ID: sonia_salman