Skip to main content
EURASIP Journal on Advances in Signal Processing

Table 3 Deconvolutional operation parameters

From: Water surface object detection using panoramic vision based on improved single-shot multibox detector

Feature layer

Input size

Kernel

Stride

Padding

Output size

Conv9_x

1*1*256

3

2

0

3*3*256

Conv8_x

3*3*256

3

2

1

5*5*256

Conv7_x

5*5*256

2

2

0

10*10*256

Conv6_x

10*10*512

3

2

1

19*19*512

Conv5_x

19*19*512

2

2

0

38*38*1024

Back to article page