EZ-CorridorKey vs Robust Video Matting

EZ-CorridorKey is a free desktop GUI for CorridorKey, an AI chroma keyer that separates foreground from background. It handles green and blue screens and detects which color is present. GVM and BiRefNet generate alpha automatically, while MatAnyone2 and VideoMaMa provide paint-and-track control. A dual viewer offers source, alpha, or composite views, an A/B wipe, pan and zoom, channel switching, and live despill, refine, and despeckle controls. The queue processes clips in sequence, with progress, sound cues, recent projects, and session saving. Exports include EXR and PNG sequences, ProRes 4444, and VP9 with alpha while preserving bit depth. Footage and project files stay on local disk, without accounts, telemetry, or cloud processing; the app does request an update check from GitHub on launch. It runs on Windows, macOS, and Linux, with Docker instructions for Linux servers or remote GPUs. Apple Silicon Macs from M1 onward are supported, though alpha generators run slower on Mac. The license is CC BY-NC-SA 4.0; donations are optional.

Robust Video Matting is a free project for separating people from video backgrounds. It processes frames with a recurrent neural network that carries temporal memory, and the project says it can perform matting in real time without additional inputs. A browser webcam demo runs the model live and can visualize its recurrent states. The conversion API accepts a video or image sequence and can produce a composited video or PNG sequence, optionally including alpha and foreground outputs.

Models are available for PyTorch, TorchHub, TorchScript, ONNX, TensorFlow, TensorFlow.js, and CoreML. MobileNetv3 is recommended for most uses; ResNet50 is larger and described as offering small performance improvements. The README reports GPU throughput figures, including 4K at 76 FPS and HD at 104 FPS on a GTX 1080 Ti, and 4K at 154 FPS and HD at 172 FPS on an RTX 3090 with FP16 under specified settings. The project cautions that file conversion runs much slower because its script does not use hardware video encoding or decoding or parallel tensor transfers. CoreML models require iOS 13 or later and do not support dynamic resolution. The code is GPL-3.0 licensed.