men1scus/birefnet
Segment foreground objects from backgrounds in images. Takes an input image (optional WxH resolution) and outputs a high...
Found 54 models (showing 21-40)
Segment foreground objects from backgrounds in images. Takes an input image (optional WxH resolution) and outputs a high...
Remove backgrounds from images. Accept an input image and return a foreground cutout with the background removed or a fo...
Segment foreground subjects and generate an alpha matte from an input image. Outputs a high-resolution saliency mask (gr...
Segment objects in images and return individual and combined masks. Uses Meta’s SAM 2 (Segment Anything v2, large varian...
Tag and segment objects in images, returning labels, bounding boxes, and pixel masks. Accepts an image as input and outp...
Segment facial regions in images. Takes an image as input and outputs color-coded semantic segmentation masks (and visua...
Detect faces in images and return face segmentation masks or transparent face cutouts. Accept a single image or a batch...
Segment anime-style images into semantic regions. Takes an image input and returns a color-coded segmentation mask image...
Remove image backgrounds from an input image and return a clean foreground cutout. Input: image. Output: image with the...
Remove image backgrounds from an input image and export the foreground mask as vector path data (SVG-style). Accepts a s...
Remove backgrounds from images of people. Input an image and get an output image with only the person retained (portrait...
Execute computer vision tasks from natural-language instructions on an input image and return an image that visualizes t...
Detect and segment watermarks in an input image, returning a binary mask highlighting watermark regions for downstream r...
Segment images into semantic classes, returning a list of labeled binary masks for each detected category. Takes a singl...
Remove backgrounds from images and return a foreground cutout with a transparent background. Takes an image as input and...
Remove and replace image backgrounds using a clean background reference shot. Provide a subject image and a background p...
Segment objects in an image and generate masks for selected objects using coordinate inputs. Remove selected objects fro...
Segment objects in an image from point prompts. Provide an image and one or more (x, y) points on the target region, and...
Generate panoptic scene graphs from an input image. Segment both “things” and “stuff” at pixel level, detect objects and...
Detect and segment specified objects in an input image and return RLE-encoded masks as JSON. Accepts an image and a comm...