Deep-Live-Cam
real time face swap and one-click video deepfake with only a single image
What is Deep-Live-Cam?
real time face swap and one-click video deepfake with only a single image
Overview of Deep-Live-Cam
What it is
Deep-Live-Cam is an open-source application for real-time face swap and one-click video deepfake using only a single image. The README describes it as a tool for the AI-generated media industry, intended to help artists animate custom characters, create content, and use models for clothing design. It includes a built-in check that prevents processing of inappropriate media such as nudity, graphic content, or sensitive material like war footage. The README also states that users should obtain consent when using a real person's face and clearly label output as a deepfake when sharing online.
Key features
- Real-time face swap driven by a single source image
- Mouth Mask, which retains the original mouth for accurate movement
- Face Mapping, to use different faces on multiple subjects simultaneously
- Watch movies with any face in real time
- Live show and performance use
- Meme creation, including a Many Faces feature
- Built-in content checks that block inappropriate media
Use cases
- Animating custom characters and creating AI-generated media content
- Running live shows and performances with a swapped face
- Watching movies with a different face applied in real time
- Creating memes, including multi-face compositions
- Using models for clothing design
Getting started
The README lists these requirements: Python (3.14 recommended; 3.11–3.14 supported), pip, git, ffmpeg, and Visual Studio 2022 Runtimes on Windows. Clone the repository, place gfpgan-1024.onnx and inswapper_128_fp16.onnx in the models folder, then install dependencies in a virtual environment.bashgit clone --depth 1 https://github.com/hacksider/Deep-Live-Cam.gitcd Deep-Live-Campython -m venv venvsource venv/bin/activatepip install -r requirements.txtpython run.pyGPU acceleration is available through execution providers: CUDA (Nvidia), CoreML (Apple Silicon), DirectML (Windows), and OpenVINO (Intel). The README notes that the first run downloads models (~300MB). Pre-built builds for Windows, Mac Silicon, CPU, NVIDIA, and AMD are also referenced.
License and pricing
Licensed under AGPL-3.0 and open source. The README also references pre-built Deep-Live-Cam 2.8 Ultimate builds with 30+ exclusive features, performance optimizations, and priority support; pricing for those builds is not stated in the provided metadata.
Based on the project's public README and repository metadata.