Real-Time Semantic Segmentation Based on Dilated Convolution Smoothing and Lightweight Up-Sampling

Cheng Xiaoyue; Zhao Longzhang; Hu Qiong; Shi Jiapeng

doi:10.3788/LOP57.021017

Laser & Optoelectronics Progress, Volume. 57, Issue 2, 21017(2020)

Real-Time Semantic Segmentation Based on Dilated Convolution Smoothing and Lightweight Up-Sampling

Cheng Xiaoyue, Zhao Longzhang, Hu Qiong, and Shi Jiapeng

Author Affiliations

College of Electrical Engineering and Control Science, Nanjing Tech University, Nanjing, Jiangsu 211816, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. Different convolution types. (a) Ordinary convolution; (b) dilated convolution; (c) dilated convolution after smoothing

Download full size

Fig. 2. Dilated convolution and smoothing effect^[24]. (a) Dilated convolution; (b) dilated convolution after smoothing

Download full size

Fig. 3. Structure of residual unit in convolution block

Download full size

Fig. 4. Structural diagram of network with HC-LUM

Download full size

Fig. 5. Comparison of segmentation accuracy between proposed segmentation method and other methods for 19 types of objects

Download full size

Fig. 6. Influence of knowledge distillation method on accuracy of each segmentation network

Download full size

Fig. 7. Comparison of segmentation accuracy for different network layers

Download full size

Fig. 8. Segmentation results of proposed method. (a) Original images 1; (b) segmentation results of original images 1 (including enlarged images of partial detail); (c) original images 2; (d) segmentation results of original images 2

Download full size

Table 1. Corresponding results of four segmentation methods on Cityscapes dataset
View table
Table 1. Corresponding results of four segmentation methods on Cityscapes dataset
Method MIOU /% Frame rate /(frame·s^-1) Parameter /10⁷
ResNeXt-18+D-Cov 72.3 36.3 1.25
ResNeXt-18+DCSM 73.7 36.1 1.39
ResNeXt-18+DCSM+HC-LUM 73.9 35.9 1.18
Proposed 76.8 40.2 1.18

Table 2. Corresponding results of four segmentation methods on CamVid dataset
View table
Table 2. Corresponding results of four segmentation methods on CamVid dataset
Method MIOU /% Frame rate /(frame·s^-1)
ResNeXt-18+D-Cov 64.2 33.9
ResNeXt-18+DCSM 64.7 33.7
ResNeXt-18+DCSM+HC-LUM 65.1 32.5
Proposed 65.3 34.2

Table 3. Comparison of proposed method with other segmentation networks (Cityscapes dataset)
View table
Table 3. Comparison of proposed method with other segmentation networks (Cityscapes dataset)
Method MIOU /% Frame rate /(frame·s^-1)
ICNet 69.5 30.3
Two-column Net 72.9 14.7
LadderDenseNet 72.82 31.0
ESPNet 60.3 112
ERFNet 68.0 11.2
GUNet 70.4 37.3
Proposed 76.8 40.2

Table 4. Comparison of proposed method with other segmentation networks (CamVid dataset)
View table
Table 4. Comparison of proposed method with other segmentation networks (CamVid dataset)
Method MIOU /% Frame rate /(frame·s^-1)
ICNet 67.1 27.8
PSPNet 69.1 5.4
Dilation10 65.3 4.4
SegNet 46.4 4.6
ERFNet 59.4 10.1
GUNet 61.8 31.3
Proposed 65.3 34.2

Tools

Get Citation

Copy Citation Text

Cheng Xiaoyue, Zhao Longzhang, Hu Qiong, Shi Jiapeng. Real-Time Semantic Segmentation Based on Dilated Convolution Smoothing and Lightweight Up-Sampling[J]. Laser & Optoelectronics Progress, 2020, 57(2): 21017

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Image Processing

Received: May. 31, 2019

Accepted: --

Published Online: Jan. 3, 2020

The Author Email:

DOI:10.3788/LOP57.021017

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Corresponding results of four segmentation methods on Cityscapes dataset

Table 1. Corresponding results of four segmentation methods on Cityscapes dataset

Table 2. Corresponding results of four segmentation methods on CamVid dataset

Table 2. Corresponding results of four segmentation methods on CamVid dataset

Table 3. Comparison of proposed method with other segmentation networks (Cityscapes dataset)

Table 3. Comparison of proposed method with other segmentation networks (Cityscapes dataset)

Table 4. Comparison of proposed method with other segmentation networks (CamVid dataset)

Table 4. Comparison of proposed method with other segmentation networks (CamVid dataset)