keypoint-rcnn-resnet50-fpn-coco-torch#

Finds people in images and maps their body joints for pose estimation and motion analysis.

Details

  • Model name: keypoint-rcnn-resnet50-fpn-coco-torch

  • Model source: https://pytorch.org/vision/main/models.html

  • Model author: Kaiming He, et al.

  • Model license: BSD 3-Clause

  • Model size: 226.05 MB

  • Exposes embeddings? no

  • Tags: keypoints, coco, torch, keypoint-rcnn, resnet, official

Requirements

  • Packages: torch, torchvision

  • CPU support

    • yes

  • GPU support

    • yes

Example usage

 1import fiftyone as fo
 2import fiftyone.zoo as foz
 3
 4dataset = foz.load_zoo_dataset(
 5    "coco-2017",
 6    split="validation",
 7    dataset_name=fo.get_default_dataset_name(),
 8    max_samples=50,
 9    shuffle=True,
10)
11
12model = foz.load_zoo_model("keypoint-rcnn-resnet50-fpn-coco-torch")
13
14dataset.apply_model(model, label_field="predictions")
15
16session = fo.launch_app(dataset)