simbrams/segformer-b5-finetuned-ade-640-640
Segment images into semantic classes. Accepts an image and returns pixel-wise masks for each class with labels and confi...
Found 55 models (showing 41-55)
Segment images into semantic classes. Accepts an image and returns pixel-wise masks for each class with labels and confi...
Remove backgrounds from images. Accepts a single image and outputs a refined foreground matte/cutout using a coarse-to-f...
Segment objects in images with fast Segment Anything–style prompting. Accepts an image plus optional text prompts (e.g.,...
Analyze documents and images from one or more image inputs plus a text prompt, returning text captions, OCR, and answers...
Segment images into foreground and background. Takes an image as input and outputs a binary mask isolating the salient o...
Segment objects in an image from text prompts and output masks. Accepts an image plus positive and negative mask prompts...
Segment images into per-object masks from an input image. Optionally guide selection with a text prompt, adjust IoU filt...
Segment objects in images into labeled masks. Accepts an image input and returns semantic segments as per-object masks w...
Generate segmentation masks for all objects in an image. Accepts a single image as input and outputs a mask image, using...
Segment interior room images into labeled regions. Accepts a single image and outputs an image with semantic segmentatio...
Perform universal image segmentation using a masked-attention mask transformer. This model processes input images to pro...
Analyze images to generate captions, extract OCR text, detect objects, and produce segmentation masks and region proposa...
Generate DensePose UV and segmentation maps from an image or video to create driving poses for MagicAnimate and other mo...
Segment DAB-stained histology images to detect protein signal and nuclei. Takes a brightfield IHC image as input and out...
Segment objects in images from a text prompt. Takes an input image and a text prompt (e.g., “person”, “car”, “clothes”)...