Run keypoints detection models from Detectron2 framework.
We strongly recommend using a virtual environment. If you're not sure where to start, we offer a tutorial here.
pip install ikomia
from ikomia.dataprocess.workflow import Workflow
from ikomia.utils.displayIO import display
# Init your workflow
wf = Workflow()
# Add keypoints detection algorithm
keypts_detector = wf.add_task(name="infer_detectron2_keypoints", auto_connect=True)
# Run the workflow on image
wf.run_on(url="https://raw.githubusercontent.com/Ikomia-hub/infer_detectron2_keypoints/main/images/rugby.jpg")
# Display result
display(keypts_detector.get_image_with_graphics(), title="Detectron2 keypoints")
Ikomia Studio offers a friendly UI with the same features as the API.
-
If you haven't started using Ikomia Studio yet, download and install it from this page.
-
For additional guidance on getting started with Ikomia Studio, check out this blog post.
from ikomia.dataprocess.workflow import Workflow
# Init your workflow
wf = Workflow()
# Add algorithm
algo = wf.add_task(name="infer_detectron2_keypoints", auto_connect=True)
algo.set_parameters({
"model_name": "COCO-Keypoints/keypoint_rcnn_R_101_FPN_3x",
"conf_det_thres": "0.5",
"conf_kp_thres": "0.05",
"cuda": "True",
"use_custom_model": "False",
"config_file": "",
"model_weight_file": "",
})
# Run the workflow on image
wf.run_on(url="https://raw.githubusercontent.com/Ikomia-hub/infer_detectron2_keypoints/main/images/rugby.jpg")
- model_name (str, default="COCO-Keypoints/keypoint_rcnn_R_101_FPN_3x"): pre-trained model name. Choose one on the list below:
- COCO-Keypoints/keypoint_rcnn_R_50_FPN_1x
- COCO-Keypoints/keypoint_rcnn_R_50_FPN_3x
- COCO-Keypoints/keypoint_rcnn_R_101_FPN_3x
- COCO-Keypoints/keypoint_rcnn_X_101_32x8d_FPN_3x
- conf_det_thres (float, default=0.5): object detection confidence.
- conf_kp_thres (float, default=0.05): keypoints detection confidence.
- cuda (bool, default=True): CUDA acceleration if True, run on CPU otherwise.
- use_custom_model (bool, default=False): flag to enable the custom train model choice.
- config_file (str, default=""): path to model config file (.yaml). Only for custom model.
- model_weight_file (str, default=""): path to model weights file (.pt). Only for custom model.
Note: parameter key and value should be in string format when added to the dictionary.
Every algorithm produces specific outputs, yet they can be explored them the same way using the Ikomia API. For a more in-depth understanding of managing algorithm outputs, please refer to the documentation.
from ikomia.dataprocess.workflow import Workflow
# Init your workflow
wf = Workflow()
# Add keypoints detection algorithm
keypts_detector = wf.add_task(name="infer_detectron2_keypoints", auto_connect=True)
# Run the workflow on image
wf.run_on(url="https://raw.githubusercontent.com/Ikomia-hub/infer_detectron2_keypoints/main/images/rugby.jpg")
# Iterate over outputs
for output in keypts_detector.get_outputs():
# Print information
print(output)
# Export it to JSON
output.to_json()
Detectron2 keypoints detection algorithm generates 2 outputs:
- Forwaded original image (CImageIO)
- Keypoints detection output (CKeypointsIO)