Real-Time 3D Hand Tracking Using GAN and RGB Camera

Implemented a real-time 3D hand tracking system using a single RGB camera, leveraging a Generative Adversarial Network (GAN) architecture. The project focused on API-level integration for efficient deployment in real-world applications. The GAN model accurately estimated 3D hand poses without the need for depth sensors or multi-camera setups. This approach demonstrated the power of generative models for vision-based human-computer interaction.