EncodeTS · February 19, 2024 06:56 · Jul 22, 2016 · Jul 22, 2016 · Jul 22, 2016 · Jul 22, 2016
diff --git a/keras VGG-Face Model.md b/keras VGG-Face Model.md
@@ -20,14 +20,15 @@ Please cite the paper if you use the models.
 ### Contents:
 
 model and usage demo: see `vgg-face-keras.py` or `vgg-face-keras-fc.py`
-The only difference between them is the last few layers,but they produce the **same result**.
+
+The only difference between them is the **last few layers**(see the code and you'll understand),but they produce the **same result**.
 
 weights: 
 
 - [Google Drive](https://drive.google.com/open?id=0B4ChsjFJvew3NkF0dTc1OGxsOFU)
 - [Baidu Yun](http://pan.baidu.com/s/1o88KALW)(password:12i0)
-- [fc Google Drive](https://drive.google.com/open?id=0B4ChsjFJvew3c2VpZWQ0eE85LVE)
-- [fc Baidu Yun](http://pan.baidu.com/s/1pKLt9LT)(password:d318)
+- [fc-version Google Drive](https://drive.google.com/open?id=0B4ChsjFJvew3c2VpZWQ0eE85LVE)
+- [fc-version Baidu Yun](http://pan.baidu.com/s/1pKLt9LT)(password:d318)
 
 ### Notice:
 Please use this model in **theano mode**.
diff --git a/vgg-face-keras-1.md → keras VGG-Face Model.md b/vgg-face-keras-1.md → keras VGG-Face Model.md
diff --git a/vgg-face-keras.md → vgg-face-keras-1.md b/vgg-face-keras.md → vgg-face-keras-1.md
diff --git a/vgg-face-keras-fc.py b/vgg-face-keras-fc.py
@@ -69,4 +69,5 @@ def vgg_face(weights_path=None):
     # Test pretrained model
     model = vgg_face('vgg-face-keras-fc.h5')
     out = model.predict(im)
-    print(out[0][0])
+    print(out[0][0])
+
diff --git a/vgg-face-keras.py b/vgg-face-keras.py
@@ -70,4 +70,5 @@ def vgg_face(weights_path=None):
     # Test pretrained model
     model = vgg_face('vgg-face-keras.h5')
     out = model.predict(im)
-    print(out[0][0])
+    print(out[0][0])
+
diff --git a/VGG-Face-keras.md → vgg-face-keras.md b/VGG-Face-keras.md → vgg-face-keras.md
diff --git a/VGG-Face-keras.md b/VGG-Face-keras.md
@@ -2,11 +2,10 @@
 
 This is the [Keras](http://keras.io/) model of [VGG-Face](http://www.robots.ox.ac.uk/~vgg/software/vgg_face/).
 
-It has been obtained through the following steps:
+It has been obtained through the following method:
 
-- export the weights of the vgg-face matconvnet model to .mat file
-- use scipy to load the weights,and convert the weight from tf mode to th mode
-- set the weights to keras model and then save the model
+- vgg-face-keras:directly convert the vgg-face matconvnet model to keras model
+- vgg-face-keras-fc:first convert vgg-face caffe model to mxnet model,and then convert it to keras model
 
 Details about the network architecture can be found in the following paper:
 
@@ -20,14 +19,15 @@ Please cite the paper if you use the models.
 
 ### Contents:
 
-model and usage demo: see `vgg-face-keras.py`
+model and usage demo: see `vgg-face-keras.py` or `vgg-face-keras-fc.py`
+The only difference between them is the last few layers,but they produce the **same result**.
 
 weights: 
 
 - [Google Drive](https://drive.google.com/open?id=0B4ChsjFJvew3NkF0dTc1OGxsOFU)
-
 - [Baidu Yun](http://pan.baidu.com/s/1o88KALW)(password:12i0)
-
+- [fc Google Drive](https://drive.google.com/open?id=0B4ChsjFJvew3c2VpZWQ0eE85LVE)
+- [fc Baidu Yun](http://pan.baidu.com/s/1pKLt9LT)(password:d318)
 
 ### Notice:
 Please use this model in **theano mode**.
diff --git a/vgg-face-keras-fc.py b/vgg-face-keras-fc.py
@@ -0,0 +1,72 @@
+from keras.models import Model
+from keras.layers import Input, Convolution2D, ZeroPadding2D, MaxPooling2D, Flatten, Dense, Dropout
+from PIL import Image
+import numpy as np
+
+def vgg_face(weights_path=None):
+    img = Input(shape=(3, 224, 224))
+
+    pad1_1 = ZeroPadding2D(padding=(1, 1))(img)
+    conv1_1 = Convolution2D(64, 3, 3, activation='relu', name='conv1_1')(pad1_1)
+    pad1_2 = ZeroPadding2D(padding=(1, 1))(conv1_1)
+    conv1_2 = Convolution2D(64, 3, 3, activation='relu', name='conv1_2')(pad1_2)
+    pool1 = MaxPooling2D((2, 2), strides=(2, 2))(conv1_2)
+
+    pad2_1 = ZeroPadding2D((1, 1))(pool1)
+    conv2_1 = Convolution2D(128, 3, 3, activation='relu', name='conv2_1')(pad2_1)
+    pad2_2 = ZeroPadding2D((1, 1))(conv2_1)
+    conv2_2 = Convolution2D(128, 3, 3, activation='relu', name='conv2_2')(pad2_2)
+    pool2 = MaxPooling2D((2, 2), strides=(2, 2))(conv2_2)
+
+    pad3_1 = ZeroPadding2D((1, 1))(pool2)
+    conv3_1 = Convolution2D(256, 3, 3, activation='relu', name='conv3_1')(pad3_1)
+    pad3_2 = ZeroPadding2D((1, 1))(conv3_1)
+    conv3_2 = Convolution2D(256, 3, 3, activation='relu', name='conv3_2')(pad3_2)
+    pad3_3 = ZeroPadding2D((1, 1))(conv3_2)
+    conv3_3 = Convolution2D(256, 3, 3, activation='relu', name='conv3_3')(pad3_3)
+    pool3 = MaxPooling2D((2, 2), strides=(2, 2))(conv3_3)
+
+    pad4_1 = ZeroPadding2D((1, 1))(pool3)
+    conv4_1 = Convolution2D(512, 3, 3, activation='relu', name='conv4_1')(pad4_1)
+    pad4_2 = ZeroPadding2D((1, 1))(conv4_1)
+    conv4_2 = Convolution2D(512, 3, 3, activation='relu', name='conv4_2')(pad4_2)
+    pad4_3 = ZeroPadding2D((1, 1))(conv4_2)
+    conv4_3 = Convolution2D(512, 3, 3, activation='relu', name='conv4_3')(pad4_3)
+    pool4 = MaxPooling2D((2, 2), strides=(2, 2))(conv4_3)
+
+    pad5_1 = ZeroPadding2D((1, 1))(pool4)
+    conv5_1 = Convolution2D(512, 3, 3, activation='relu', name='conv5_1')(pad5_1)
+    pad5_2 = ZeroPadding2D((1, 1))(conv5_1)
+    conv5_2 = Convolution2D(512, 3, 3, activation='relu', name='conv5_2')(pad5_2)
+    pad5_3 = ZeroPadding2D((1, 1))(conv5_2)
+    conv5_3 = Convolution2D(512, 3, 3, activation='relu', name='conv5_3')(pad5_3)
+    pool5 = MaxPooling2D((2, 2), strides=(2, 2))(conv5_3)
+
+    flat = Flatten()(pool5)
+    fc6 = Dense(4096, activation='relu', name='fc6')(flat)
+    fc6_drop = Dropout(0.5)(fc6)
+    fc7 = Dense(4096, activation='relu', name='fc7')(fc6_drop)
+    fc7_drop = Dropout(0.5)(fc7)
+    out = Dense(2622, activation='softmax', name='fc8')(fc7_drop)
+
+    model = Model(input=img, output=out)
+
+    if weights_path:
+        model.load_weights(weights_path)
+
+    return model
+
+if __name__ == "__main__":
+    im = Image.open('A.J._Buckley.jpg')
+    im = im.resize((224,224))
+    im = np.array(im).astype(np.float32)
+    #    im[:,:,0] -= 129.1863
+    #    im[:,:,1] -= 104.7624
+    #    im[:,:,2] -= 93.5940
+    im = im.transpose((2,0,1))
+    im = np.expand_dims(im, axis=0)
+
+    # Test pretrained model
+    model = vgg_face('vgg-face-keras-fc.h5')
+    out = model.predict(im)
+    print(out[0][0])
diff --git a/VGG-Face-keras.py → vgg-face-keras.py b/VGG-Face-keras.py → vgg-face-keras.py
@@ -42,7 +42,6 @@ def vgg_face(weights_path=None):
     conv5_3 = Convolution2D(512, 3, 3, activation='relu', name='conv5_3')(pad5_3)
     pool5 = MaxPooling2D((2, 2), strides=(2, 2))(conv5_3)
 
-
     fc6 = Convolution2D(4096, 7, 7, activation='relu', name='fc6')(pool5)
     fc6_drop = Dropout(0.5)(fc6)
     fc7 = Convolution2D(4096, 1, 1, activation='relu', name='fc7')(fc6_drop)

diff --git a/VGG-Face-keras.py b/VGG-Face-keras.py
@@ -71,4 +71,4 @@ def vgg_face(weights_path=None):
     # Test pretrained model
     model = vgg_face('vgg-face-keras.h5')
     out = model.predict(im)
-    print(np.argmax(out))
+    print(out[0][0])
diff --git a/VGG-Face-keras.md b/VGG-Face-keras.md
@@ -0,0 +1,33 @@
+## VGG-Face model for Keras
+
+This is the [Keras](http://keras.io/) model of [VGG-Face](http://www.robots.ox.ac.uk/~vgg/software/vgg_face/).
+
+It has been obtained through the following steps:
+
+- export the weights of the vgg-face matconvnet model to .mat file
+- use scipy to load the weights,and convert the weight from tf mode to th mode
+- set the weights to keras model and then save the model
+
+Details about the network architecture can be found in the following paper:
+
+```
+Deep Face Recognition
+O. M. Parkhi, A. Vedaldi, A. Zisserman
+British Machine Vision Conference, 2015
+```
+
+Please cite the paper if you use the models.
+
+### Contents:
+
+model and usage demo: see `vgg-face-keras.py`
+
+weights: 
+
+- [Google Drive](https://drive.google.com/open?id=0B4ChsjFJvew3NkF0dTc1OGxsOFU)
+
+- [Baidu Yun](http://pan.baidu.com/s/1o88KALW)(password:12i0)
+
+
+### Notice:
+Please use this model in **theano mode**.
diff --git a/VGG-Face-keras.py b/VGG-Face-keras.py
@@ -0,0 +1,74 @@
+from keras.models import Model
+from keras.layers import Input, Convolution2D, ZeroPadding2D, MaxPooling2D, Flatten, Dropout, Activation
+from PIL import Image
+import numpy as np
+
+def vgg_face(weights_path=None):
+    img = Input(shape=(3, 224, 224))
+
+    pad1_1 = ZeroPadding2D(padding=(1, 1))(img)
+    conv1_1 = Convolution2D(64, 3, 3, activation='relu', name='conv1_1')(pad1_1)
+    pad1_2 = ZeroPadding2D(padding=(1, 1))(conv1_1)
+    conv1_2 = Convolution2D(64, 3, 3, activation='relu', name='conv1_2')(pad1_2)
+    pool1 = MaxPooling2D((2, 2), strides=(2, 2))(conv1_2)
+
+    pad2_1 = ZeroPadding2D((1, 1))(pool1)
+    conv2_1 = Convolution2D(128, 3, 3, activation='relu', name='conv2_1')(pad2_1)
+    pad2_2 = ZeroPadding2D((1, 1))(conv2_1)
+    conv2_2 = Convolution2D(128, 3, 3, activation='relu', name='conv2_2')(pad2_2)
+    pool2 = MaxPooling2D((2, 2), strides=(2, 2))(conv2_2)
+
+    pad3_1 = ZeroPadding2D((1, 1))(pool2)
+    conv3_1 = Convolution2D(256, 3, 3, activation='relu', name='conv3_1')(pad3_1)
+    pad3_2 = ZeroPadding2D((1, 1))(conv3_1)
+    conv3_2 = Convolution2D(256, 3, 3, activation='relu', name='conv3_2')(pad3_2)
+    pad3_3 = ZeroPadding2D((1, 1))(conv3_2)
+    conv3_3 = Convolution2D(256, 3, 3, activation='relu', name='conv3_3')(pad3_3)
+    pool3 = MaxPooling2D((2, 2), strides=(2, 2))(conv3_3)
+
+    pad4_1 = ZeroPadding2D((1, 1))(pool3)
+    conv4_1 = Convolution2D(512, 3, 3, activation='relu', name='conv4_1')(pad4_1)
+    pad4_2 = ZeroPadding2D((1, 1))(conv4_1)
+    conv4_2 = Convolution2D(512, 3, 3, activation='relu', name='conv4_2')(pad4_2)
+    pad4_3 = ZeroPadding2D((1, 1))(conv4_2)
+    conv4_3 = Convolution2D(512, 3, 3, activation='relu', name='conv4_3')(pad4_3)
+    pool4 = MaxPooling2D((2, 2), strides=(2, 2))(conv4_3)
+
+    pad5_1 = ZeroPadding2D((1, 1))(pool4)
+    conv5_1 = Convolution2D(512, 3, 3, activation='relu', name='conv5_1')(pad5_1)
+    pad5_2 = ZeroPadding2D((1, 1))(conv5_1)
+    conv5_2 = Convolution2D(512, 3, 3, activation='relu', name='conv5_2')(pad5_2)
+    pad5_3 = ZeroPadding2D((1, 1))(conv5_2)
+    conv5_3 = Convolution2D(512, 3, 3, activation='relu', name='conv5_3')(pad5_3)
+    pool5 = MaxPooling2D((2, 2), strides=(2, 2))(conv5_3)
+
+
+    fc6 = Convolution2D(4096, 7, 7, activation='relu', name='fc6')(pool5)
+    fc6_drop = Dropout(0.5)(fc6)
+    fc7 = Convolution2D(4096, 1, 1, activation='relu', name='fc7')(fc6_drop)
+    fc7_drop = Dropout(0.5)(fc7)
+    fc8 = Convolution2D(2622, 1, 1, name='fc8')(fc7_drop)
+    flat = Flatten()(fc8)
+    out = Activation('softmax')(flat)
+
+    model = Model(input=img, output=out)
+
+    if weights_path:
+        model.load_weights(weights_path)
+
+    return model
+
+if __name__ == "__main__":
+    im = Image.open('A.J._Buckley.jpg')
+    im = im.resize((224,224))
+    im = np.array(im).astype(np.float32)
+#    im[:,:,0] -= 129.1863
+#    im[:,:,1] -= 104.7624
+#    im[:,:,2] -= 93.5940
+    im = im.transpose((2,0,1))
+    im = np.expand_dims(im, axis=0)
+
+    # Test pretrained model
+    model = vgg_face('vgg-face-keras.h5')
+    out = model.predict(im)
+    print(np.argmax(out))