MediaPipe Selfie Segmenter
Coming soonSeparate the prominent person from an image background. Returns a transparent PNG and an alpha-mask image for replacing or removing portrait backgrounds.
- Repository created
- Jun 13, 2019
One input, different models. Compare outputs and measured speed side by side on your own device.
Model comparisons are coming with the upgraded LocalAI Tools. Explore the models below, or join the waitlist to hear when it’s ready.
Separate the prominent person from an image background. Returns a transparent PNG and an alpha-mask image for replacing or removing portrait backgrounds.
Remove image backgrounds locally with the compact BiRefNet Lite 512 browser export.
Separate people from an image background for background removal or replacement. Returns a foreground matte and transparent-image output; designed primarily for human subjects.
Remove a portrait background by estimating how much each pixel belongs to the person. Returns a soft foreground matte without requiring a manually prepared trimap; fine hair and edges may need review.
Remove still-image backgrounds with the original BEN2 Base network at1024×1024 in an isolated browser worker.
Remove still-image backgrounds using the MVANet multi-view segmentation checkpoint in a module worker.