Sheffield Softworks Keyper is a people keyer that uses Machine Learning to extract people from their backgrounds without requiring a green screen or prepared backdrops. It runs in After Effects, Premiere Pro and DaVinci Resolve, as well as Final Cut Pro X and Motion. Rather than replacing traditional green screens used in special effects shots, Keyper lets you apply filters and effects separately to the people and/or the backgrounds, creating effects such as 3D still photos and the parallax effect.
People are not tracked between frames — each frame is analyzed on its own — and all people recognized by Apple's Machine Learning are separated from the background. Version 2.0 adds keyframeable preprocessing for brightness, saturation, contrast, sharpness and LUT loading that does not alter the output image, along with matte processing modes (Full, Outline, Inline, Core, Combo), a Colored Matte output with Alpha, and a Patch control.
Version 2.1 adds a Subject detector for macOS 14 Sonoma and later that identifies human and non-human subjects. Keyper requires Noise Industries FxFactory to run and works on Apple Silicon and Intel processors.