deepid model added

deepface/DeepFace.py

@@ -15,7 +15,7 @@ import tensorflow as tf
 import pickle
 
 from deepface import DeepFace
-from deepface.basemodels import VGGFace, OpenFace, Facenet, FbDeepFace
+from deepface.basemodels import VGGFace, OpenFace, Facenet, FbDeepFace, DeepID
 from deepface.extendedmodels import Age, Gender, Race, Emotion
 from deepface.commons import functions, realtime, distance as dst
 
@@ -196,6 +196,10 @@ def verify(img1_path, img2_path=''
 			print("Using FB DeepFace model backend", distance_metric,"distance.")
 			model = FbDeepFace.loadModel()
 		
+		if model_name == 'DeepID':
+			print("Using DeepID2 model backend", distance_metric,"distance.")
+			model = DeepID.loadModel()
+
 		else:
 			raise ValueError("Invalid model_name passed - ", model_name)
 	else: #model != None
@@ -205,6 +209,9 @@ def verify(img1_path, img2_path=''
 	#face recognition models have different size of inputs
 	input_shape = model.layers[0].input_shape[1:3]
 	
+	input_shape_x = input_shape[0]
+	input_shape_y = input_shape[1]
+
 	#------------------------------
 
 	#tuned thresholds for model and metric pair
@@ -225,8 +232,8 @@ def verify(img1_path, img2_path=''
 			#----------------------
 			#crop and align faces
 
-			img1 = functions.detectFace(img1_path, input_shape, enforce_detection = enforce_detection)
-			img2 = functions.detectFace(img2_path, input_shape, enforce_detection = enforce_detection)
+			img1 = functions.detectFace(img1_path, (input_shape_y, input_shape_x), enforce_detection = enforce_detection)
+			img2 = functions.detectFace(img2_path, (input_shape_y, input_shape_x), enforce_detection = enforce_detection)
 
 			#----------------------
 			#find embeddings
@@ -499,9 +506,13 @@ def find(img_path, db_path
 			elif model_name == 'DeepFace':
 				print("Using FB DeepFace model backend", distance_metric,"distance.")
 				model = FbDeepFace.loadModel()
+			elif model_name == 'DeepID':
+				print("Using DeepID model backend", distance_metric,"distance.")
+				model = DeepID.loadModel()
 			elif model_name == 'Ensemble':
 				
 				print("Ensemble learning enabled")
+				#TODO: include DeepID in ensemble method
 				
 				import lightgbm as lgb #lightgbm==2.3.1
 				
@@ -585,7 +596,9 @@ def find(img_path, db_path
 				if model_name != 'Ensemble':
 				
 					input_shape = model.layers[0].input_shape[1:3]
-					img = functions.detectFace(employee, input_shape, enforce_detection = enforce_detection)
+					input_shape_x = input_shape[0]; input_shape_y = input_shape[1]
+					
+					img = functions.detectFace(employee, (input_shape_y, input_shape_x), enforce_detection = enforce_detection)
 					representation = model.predict(img)[0,:]
 					
 					instance = []
@@ -600,7 +613,9 @@ def find(img_path, db_path
 					for j in model_names:
 						model = models[j]
 						input_shape = model.layers[0].input_shape[1:3]
-						img = functions.detectFace(employee, input_shape, enforce_detection = enforce_detection)
+						input_shape_x = input_shape[0]; input_shape_y = input_shape[1]
+						
+						img = functions.detectFace(employee, (input_shape_y, input_shape_x), enforce_detection = enforce_detection)
 						representation = model.predict(img)[0,:]
 						instance.append(representation)
 				
@@ -705,7 +720,9 @@ def find(img_path, db_path
 			
 			if model_name != 'Ensemble':
 				input_shape = model.layers[0].input_shape[1:3]
-				img = functions.detectFace(img_path, input_shape, enforce_detection = enforce_detection)
+				input_shape_x = input_shape[0]; input_shape_y = input_shape[1]
+				
+				img = functions.detectFace(img_path, (input_shape_y, input_shape_x), enforce_detection = enforce_detection)
 				target_representation = model.predict(img)[0,:]
 		
 				distances = []

deepface/basemodels/DeepID.py

@@ -0,0 +1,52 @@
+import os
+from pathlib import Path
+import gdown
+import keras
+from keras.models import Model
+from keras.layers import Conv2D, Activation, Input, Add, MaxPooling2D, Flatten, Dense, Dropout
+import zipfile
+
+#-------------------------------------
+
+def loadModel():
+	
+	myInput = Input(shape=(55, 47, 3))
+	
+	x = Conv2D(20, (4, 4), name='Conv1', activation='relu', input_shape=(55, 47, 3))(myInput)
+	x = MaxPooling2D(pool_size=2, strides=2, name='Pool1')(x)
+	x = Dropout(rate=1, name='D1')(x)
+	
+	x = Conv2D(40, (3, 3), name='Conv2', activation='relu')(x)
+	x = MaxPooling2D(pool_size=2, strides=2, name='Pool2')(x)
+	x = Dropout(rate=1, name='D2')(x)
+	
+	x = Conv2D(60, (3, 3), name='Conv3', activation='relu')(x)
+	x = MaxPooling2D(pool_size=2, strides=2, name='Pool3')(x)
+	x = Dropout(rate=1, name='D3')(x)
+	
+	x1 = Flatten()(x)
+	fc11 = Dense(160, name = 'fc11')(x1)
+	
+	x2 = Conv2D(80, (2, 2), name='Conv4', activation='relu')(x)
+	x2 = Flatten()(x2)
+	fc12 = Dense(160, name = 'fc12')(x2)
+	
+	y = Add()([fc11, fc12])
+	y = Activation('relu', name = 'deepid')(y)
+	
+	model = Model(inputs=[myInput], outputs=y)
+
+	#---------------------------------
+	
+	home = str(Path.home())
+	
+	if os.path.isfile(home+'/.deepface/weights/deepid_keras_weights.h5') != True:
+		print("deepid_keras_weights.h5 will be downloaded...")
+		
+		url = 'https://drive.google.com/uc?id=1uRLtBCTQQAvHJ_KVrdbRJiCKxU8m5q2J'
+		output = home+'/.deepface/weights/deepid_keras_weights.h5'
+		gdown.download(url, output, quiet=False)
+		
+	model.load_weights(home+'/.deepface/weights/deepid_keras_weights.h5')	
+	
+	return model

deepface/commons/functions.py

@@ -129,6 +129,14 @@ def findThreshold(model_name, distance_metric):
 		elif distance_metric == 'euclidean_l2':
 			threshold = 0.64
 	
+	elif model_name == 'DeepID':
+		if distance_metric == 'cosine':
+			threshold = 0.015
+		elif distance_metric == 'euclidean':
+			threshold = 45
+		elif distance_metric == 'euclidean_l2':
+			threshold = 0.17
+	
 	return threshold
 
 def get_opencv_path():

tests/unit_tests.py

@@ -105,7 +105,8 @@ dataset = [
 	['dataset/img6.jpg', 'dataset/img9.jpg', False],
 ]
 
-models = ['VGG-Face', 'Facenet', 'OpenFace', 'DeepFace']
+models = ['DeepID']
+#models = ['VGG-Face', 'Facenet', 'OpenFace', 'DeepFace', 'DeepID']
 metrics = ['cosine', 'euclidean', 'euclidean_l2']
 
 passed_tests = 0; test_cases = 0
@@ -178,3 +179,4 @@ facial_attribute_models["gender"] = gender_model
 facial_attribute_models["race"] = race_model
 
 resp_obj = DeepFace.analyze("dataset/img1.jpg", models=facial_attribute_models)
+