Detectors World

Hand, pose and face detection in a few lines of Python. Detectors World gives you ready-made computer vision detectors that work directly with OpenCV images.

Get started Browse examples

pip install detectors-world

import cv2 as cv
from detectors_world import create_detector

cap = cv.VideoCapture(0)

with create_detector("hand") as hand:
    while True:
        ok, frame = cap.read()
        if not ok:
            break
        hands = hand.detect(frame)  # draws on frame and returns what it found
        cv.imshow("Hand Detection", frame)
        if cv.waitKey(1) & 0xFF == ord("q"):  # press q to quit
            break

cap.release()
cv.destroyAllWindows()

Detectors

• Hand

  ---

  

  Finds hands, 21 landmarks each, and tells left from right.

• Pose

  ---

  

  Finds bodies and 33 landmarks from head to toe.

• Face

  ---

  

  Finds face boxes, confidence scores and 6 key points.

• Face Mesh

  ---

  

  Finds 478 landmarks covering the face, lips, eyes and irises.

Why Detectors World

• Short code

  One call creates a detector, and one call runs it. Models download by themselves the first time.

• Made for OpenCV

  Pass the frame you already have. Drawing happens on it directly, so cv.imshow shows the result.

• Simple results

  Get pixel positions you can compare, measure and draw with, no conversion needed.

• Complete reference

  Every landmark is mapped and named in the landmark maps.