segformer-b0-ade20k-torch

Efficient transformer-based semantic segmentation model for scene parsing with 150 classes.

Details

• Model name: segformer-b0-ade20k-torch

• Model source: https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512

• Model author: Enze Xie, et al.

• Model license: MIT

• Model size: 14.32 MB

• Exposes embeddings? no

• Tags: segmentation, torch, segformer, 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("segformer-b0-ade20k-torch")

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

session = fo.launch_app(dataset)