metadata

license: cc-by-4.0
library_name: ultralytics
pipeline_tag: object-detection
inference: false
tags:
  - roboflow
  - hand_gesture

Yolov8x Tuned to Hand Gestures

Yolo v8x Finetuned on the hand gestures roboflow dataset.

Inference

from ultralyticsplus import YOLO, render_result

# load model
model = YOLO('lewiswatson/yolov8x-tuned-hand-gestures')

# set image
image = 'https://thumbs.dreamstime.com/b/young-man-showing-peace-sign-isolated-over-yellow-background-133210696.jpg'

# perform inference
results = model.predict(image)

# observe results
print(results[0].boxes)
render = render_result(model=model, image=image, result=results[0])
render.show()

Training Parameters

Trained using Ultralytics and Ultralyticsplus.

Epochs: 10
Optimiser: SGD
Patience 50
Imgsz: 640
Batch: -1