Improving Faster R-CNN Framework for Fast Vehicle Detection

Table 3

The architecture of the classifier.

Type/stride	Filter shape	Input size
Conv dw/s2	3 × 3 × 512 dw	14 × 14 × 512
Conv/s1	1 × 1 × 512 × 1024	7 × 7 × 512
Conv dw/s2	3 × 3 × 1024 dw	7 × 7 × 1024
Conv/s1	1 × 1 × 1024 × 1024	7 × 7 × 1024
Avg pool/s1	Pool 7 × 7	7 × 7 × 1024
FC/s1	1024 × 2	1 × 1 × 1024
Softmax	Classification	RoI × 2
FC/s1	1024 × 4	1 × 1 × 1024
Linear	Regression	RoI × 4