test_image_file = []
test_folders = [
'SingleOneTest', 'SingleTwoTest', 'SingleThreeTest', 'SingleFourTest',
'SingleFiveTest', 'SingleSixTest', 'SingleSevenTest', 'SingleEightTest'
]
for folder in test_folders:
test_image_file = test_image_file + os.listdir('../EgoGesture Dataset/' +
folder + '/')
print('# of images for performance analysis: ', len(test_image_file))
""" Key points Detection """
model = model()
model.summary()
model.load_weights('weights/performance.h5')
def classify(image):
image = np.asarray(image)
image = cv2.resize(image, (128, 128))
image = image.astype('float32')
image = image / 255.0
image = np.expand_dims(image, axis=0)
probability, position = model.predict(image)
probability = probability[0]
position = position[0]
return probability, position
def class_finder(prob):
import time
import numpy as np
from statistics import mean
from net.network import model
from preprocess.label_gen_test import label_generator_testset
test_image_file = []
test_folders = ['SingleOneTest', 'SingleTwoTest', 'SingleThreeTest', 'SingleFourTest', 'SingleFiveTest']
for folder in test_folders:
test_image_file = test_image_file + os.listdir('../EgoGesture Dataset/' + folder + '/')
""" Key points Detection """
model = model()
model.summary()
model.load_weights('weights/comparison.h5')
def classify(image):
image = np.asarray(image)
image = cv2.resize(image, (128, 128))
image = image.astype('float32')
image = image / 255.0
image = np.expand_dims(image, axis=0)
probability, position = model.predict(image)
probability = probability[0]
position = position[0]
return probability, position
def class_finder(prob):