OpenCV类Tensorflow 2.3.0 ResNet模型的冻结与优化

Question

我有一个ResNet类型的模型，我已经简化了它，我想与OpenCV一起使用

Layer (type)                    Output Shape         Param #     Connected to                     
==================================================================================================
input_1 (InputLayer)            [(None, 50, 50, 1)]  0                                            
__________________________________________________________________________________________________
batch_normalization (BatchNorma (None, 50, 50, 1)    4           input_1[0][0]                    
__________________________________________________________________________________________________
conv2d (Conv2D)                 (None, 50, 50, 20)   200         batch_normalization[0][0]        
__________________________________________________________________________________________________
re_lu (ReLU)                    (None, 50, 50, 20)   0           conv2d[0][0]                     
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 50, 50, 20)   80          re_lu[0][0]                      
__________________________________________________________________________________________________

...

conv2d_13 (Conv2D)              (None, 13, 13, 80)   57680       batch_normalization_12[0][0]     
__________________________________________________________________________________________________
add_5 (Add)                     (None, 13, 13, 80)   0           conv2d_14[0][0]                  
                                                                 conv2d_13[0][0]                  
__________________________________________________________________________________________________
re_lu_12 (ReLU)                 (None, 13, 13, 80)   0           add_5[0][0]                      
__________________________________________________________________________________________________
batch_normalization_13 (BatchNo (None, 13, 13, 80)   320         re_lu_12[0][0]                   
__________________________________________________________________________________________________
average_pooling2d (AveragePooli (None, 3, 3, 80)     0           batch_normalization_13[0][0]     
__________________________________________________________________________________________________
flatten (Flatten)               (None, 720)          0           average_pooling2d[0][0]          
__________________________________________________________________________________________________
dense (Dense)                   (None, 2)            1442        flatten[0][0]                    
==================================================================================================

我按照教程冻结模型，并使用Tensorflow 1.5tensorflow.python.tools.optimize_for_inference工具对其进行优化

• https://medium.com/@sebastingarcaacosta/how-to-export-a-tensorflow-2-x-keras-model-to-a-frozen-and-optimized-graph-39740846d9eb

所有的步骤都很好，现在我可以用OpenCV加载我的模型了

net = cv2.dnn.readNetFromTensorflow("optmized_graph.pb", "optmized_graph.pbtxt")
# or
net = cv2.dnn.readNet("optmized_graph.pb", "optmized_graph.pbtxt")

但是当我试图得到预测时，我得到了一些奇怪的错误：

image = np.random.rand(1, 50, 50).astype("float32")
net.setInput(image, 'x')

# or

image = np.random.rand(1, 50, 50, 1).astype("float32")
net.setInput(image, 'x')

错误是

[ERROR:0] global /tmp/pip-req-build-6amqbhlx/opencv/modules/dnn/src/dnn.cpp (3441) getLayerShapesRecursively OPENCV/DNN: [Convolution]:(mask-try4/conv2d/Conv2D): ge
tMemoryShapes() throws exception. inputs=1 outputs=0/1 blobs=2
[ERROR:0] global /tmp/pip-req-build-6amqbhlx/opencv/modules/dnn/src/dnn.cpp (3447) getLayerShapesRecursively     input[0] = [ 1 50 ]
[ERROR:0] global /tmp/pip-req-build-6amqbhlx/opencv/modules/dnn/src/dnn.cpp (3455) getLayerShapesRecursively     blobs[0] = CV_32FC1 [ 20 1 3 3 ]
[ERROR:0] global /tmp/pip-req-build-6amqbhlx/opencv/modules/dnn/src/dnn.cpp (3455) getLayerShapesRecursively     blobs[1] = CV_32FC1 [ 20 1 ]
[ERROR:0] global /tmp/pip-req-build-6amqbhlx/opencv/modules/dnn/src/dnn.cpp (3457) getLayerShapesRecursively Exception message: OpenCV(4.4.0) /tmp/pip-req-build-6am
qbhlx/opencv/modules/dnn/src/layers/convolution_layer.cpp:348: error: (-215:Assertion failed) ngroups > 0 && inpCn % ngroups == 0 && outCn % ngroups == 0 in functio
n 'getMemoryShapes'

Traceback (most recent call last):
  File "test_with_opencv.py", line 17, in <module>
    detections = net.forward()
cv2.error: OpenCV(4.4.0) /tmp/pip-req-build-6amqbhlx/opencv/modules/dnn/src/layers/convolution_layer.cpp:348: error: (-215:Assertion failed) ngroups > 0 && inpCn %
ngroups == 0 && outCn % ngroups == 0 in function 'getMemoryShapes'

我的问题是：

• 我做错了什么
• 如何克服错误并使模型工作
• 我的错误消息中的形状是input[0] = [ 1 50 ]？（我经过(1,50,50)）

版本：

• Tensorflow:2.3.0
• OpenCV:4.4.0.44
• Python:3.6.9