group-vit-segmentation-transformer-torch

Hugging Face Transformers model for zero-shot semantic segmentation.

Details

• Model name: group-vit-segmentation-transformer-torch

• Model source: https://huggingface.co/docs/transformers/en/tasks/mask_generation

• Model author: Thomas Wolf, et al.

• Model license: Apache 2.0

• Model size: 212.80 MB

• Exposes embeddings? yes

• Tags: segmentation, embeddings, torch, transformers, zero-shot, official

Requirements

• Packages: torch, torchvision, transformers

• CPU support

  • yes

• GPU support

  • yes

Example usage

import fiftyone as fo
import fiftyone.zoo as foz

dataset = foz.load_zoo_dataset(
    "coco-2017",
    split="validation",
    dataset_name=fo.get_default_dataset_name(),
    max_samples=50,
    shuffle=True,
)

model = foz.load_zoo_model("group-vit-segmentation-transformer-torch",
    text_prompt="A photo of a",
    classes=["person", "dog", "cat", "bird", "car", "tree", "other"])

dataset.apply_model(model, label_field="predictions")

session = fo.launch_app(dataset)