mediapipe-gesture_recognizer.py

import mediapipe as mp # type: ignore

BaseOptions = mp.tasks.BaseOptions
GestureRecognizer = mp.tasks.vision.GestureRecognizer
GestureRecognizerOptions = mp.tasks.vision.GestureRecognizerOptions
VisionRunningMode = mp.tasks.vision.RunningMode

options =GestureRecognizerOptions(
    base_options=BaseOptions(model_asset_path='gesture_recognizer.task'),
    running_mode=VisionRunningMode.VIDEO)
recognizer = GestureRecognizer.create_from_options(options)

import cv2

def opencv_to_mediapipe_image(cv_image):
    return mp.Image(image_format=mp.ImageFormat.SRGB, data=cv_image)

cap = cv2.VideoCapture(0)
font = cv2.FONT_HERSHEY_SIMPLEX
fps = cap.get(cv2.CAP_PROP_FPS)

timestamps = [cap.get(cv2.CAP_PROP_POS_MSEC)]
calc_timestamps = [0.0]

while(cap.isOpened()):
    ret,frame = cap.read()
    if ret:
        timestamps.append(cap.get(cv2.CAP_PROP_POS_MSEC))
        calc_timestamps.append(calc_timestamps[-1] + 1000/fps)

        recognition_result =  recognizer.recognize_for_video(opencv_to_mediapipe_image(frame), int(calc_timestamps[-1]))
        try:
            top_gesture = recognition_result.gestures[0][0].category_name
            frame = cv2.putText(frame, str(top_gesture), (10, int(cap.get(4)-10)), font, 1, (255, 0, 0), 5, cv2.LINE_AA)
        except:
            pass

        f = open("test.txt", "a")
        f.write(str(recognition_result))
        f.close()

        cv2.imshow('frame',frame)
        if cv2.waitKey(1) & 0xFF == ord('q'):
            break
    else:
        break
cap.release()
cv2.destroyAllWindows()