Faster RCNN model summary for the object detection
Model: "model_5"
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_1 (InputLayer) [(None, 400, 300, 3) 0
__________________________________________________________________________________________________
block1_conv1 (Conv2D) (None, 400, 300, 64) 1792 input_1[0][0]
__________________________________________________________________________________________________
block1_conv2 (Conv2D) (None, 400, 300, 64) 36928 block1_conv1[0][0]
__________________________________________________________________________________________________
block1_pool (MaxPooling2D) (None, 200, 150, 64) 0 block1_conv2[0][0]
__________________________________________________________________________________________________
block2_conv1 (Conv2D) (None, 200, 150, 128 73856 block1_pool[0][0]
__________________________________________________________________________________________________
block2_conv2 (Conv2D) (None, 200, 150, 128 147584 block2_conv1[0][0]
__________________________________________________________________________________________________
block2_pool (MaxPooling2D) (None, 100, 75, 128) 0 block2_conv2[0][0]
__________________________________________________________________________________________________
block3_conv1 (Conv2D) (None, 100, 75, 256) 295168 block2_pool[0][0]
__________________________________________________________________________________________________
block3_conv2 (Conv2D) (None, 100, 75, 256) 590080 block3_conv1[0][0]
__________________________________________________________________________________________________
block3_conv3 (Conv2D) (None, 100, 75, 256) 590080 block3_conv2[0][0]
__________________________________________________________________________________________________
block3_pool (MaxPooling2D) (None, 50, 37, 256) 0 block3_conv3[0][0]
__________________________________________________________________________________________________
block4_conv1 (Conv2D) (None, 50, 37, 512) 1180160 block3_pool[0][0]
__________________________________________________________________________________________________
block4_conv2 (Conv2D) (None, 50, 37, 512) 2359808 block4_conv1[0][0]
__________________________________________________________________________________________________
block4_conv3 (Conv2D) (None, 50, 37, 512) 2359808 block4_conv2[0][0]
__________________________________________________________________________________________________
block4_pool (MaxPooling2D) (None, 25, 18, 512) 0 block4_conv3[0][0]
__________________________________________________________________________________________________
block5_conv1 (Conv2D) (None, 25, 18, 512) 2359808 block4_pool[0][0]
__________________________________________________________________________________________________
block5_conv2 (Conv2D) (None, 25, 18, 512) 2359808 block5_conv1[0][0]
__________________________________________________________________________________________________
block5_conv3 (Conv2D) (None, 25, 18, 512) 2359808 block5_conv2[0][0]
__________________________________________________________________________________________________
input_2 (InputLayer) [(None, None, 4)] 0
__________________________________________________________________________________________________
roi_pooling_conv (RoiPoolingCon (1, 4, 7, 7, 512) 0 block5_conv3[0][0]
input_2[0][0]
__________________________________________________________________________________________________
time_distributed (TimeDistribut (1, 4, 25088) 0 roi_pooling_conv[0][0]
__________________________________________________________________________________________________
time_distributed_1 (TimeDistrib (1, 4, 4096) 102764544 time_distributed[0][0]
__________________________________________________________________________________________________
time_distributed_2 (TimeDistrib (1, 4, 4096) 0 time_distributed_1[0][0]
__________________________________________________________________________________________________
time_distributed_3 (TimeDistrib (1, 4, 4096) 16781312 time_distributed_2[0][0]
__________________________________________________________________________________________________
rpn_conv1 (Conv2D) (None, 25, 18, 512) 2359808 block5_conv3[0][0]
__________________________________________________________________________________________________
time_distributed_4 (TimeDistrib (1, 4, 4096) 0 time_distributed_3[0][0]
__________________________________________________________________________________________________
rpn_out_class (Conv2D) (None, 25, 18, 9) 4617 rpn_conv1[0][0]
__________________________________________________________________________________________________
rpn_out_regress (Conv2D) (None, 25, 18, 36) 18468 rpn_conv1[0][0]
__________________________________________________________________________________________________
dense_class_38 (TimeDistributed (1, 4, 38) 155686 time_distributed_4[0][0]
__________________________________________________________________________________________________
dense_regress_38 (TimeDistribut (1, 4, 148) 606356 time_distributed_4[0][0]
==================================================================================================
Total params: 137,405,479
Trainable params: 137,405,479
Non-trainable params: 0
__________________________________________________________________________________________________