Deep-Live-Cam

Why we included this project

Deep-Live-Cam swaps a face onto a live webcam feed or an existing video in real time, using just a single reference image. You can keep the original mouth for accurate lip movement and map different faces onto multiple people at once. It runs on CPU for basic use and accelerates on NVIDIA CUDA or Apple Silicon CoreML, so most machines can handle it. The project is upfront about the risks: built-in checks refuse to process nudity or graphic content, and users are asked to label deepfake output when sharing. That matters if you are evaluating it for animation, character work, or performance. Manual setup takes real technical skill, though prebuilt quickstart builds offer a faster path.