Skip to content

An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Standard

An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling. / Dong, Jiansheng; Yuan, Jingling; Li, Lin; Zhong, Xian; Liu, Weiru.

31st International Conference on Tools with Artificial Intelligence (ICTAI2019). Institute of Electrical and Electronics Engineers (IEEE), 2019.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Harvard

Dong, J, Yuan, J, Li, L, Zhong, X & Liu, W 2019, An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling. in 31st International Conference on Tools with Artificial Intelligence (ICTAI2019). Institute of Electrical and Electronics Engineers (IEEE), International Conference on Tools with Artificial Intelligence, Portland, United States, 4/11/19.

APA

Dong, J., Yuan, J., Li, L., Zhong, X., & Liu, W. (Accepted/In press). An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling. In 31st International Conference on Tools with Artificial Intelligence (ICTAI2019) Institute of Electrical and Electronics Engineers (IEEE).

Vancouver

Dong J, Yuan J, Li L, Zhong X, Liu W. An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling. In 31st International Conference on Tools with Artificial Intelligence (ICTAI2019). Institute of Electrical and Electronics Engineers (IEEE). 2019

Author

Dong, Jiansheng ; Yuan, Jingling ; Li, Lin ; Zhong, Xian ; Liu, Weiru. / An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling. 31st International Conference on Tools with Artificial Intelligence (ICTAI2019). Institute of Electrical and Electronics Engineers (IEEE), 2019.

Bibtex

@inproceedings{6700ce6a280a4ed5a829cd1c84e8fb4e,
title = "An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling",
abstract = "—Efficient and accurate semantic segmentation is particularly important especially for applications like autonomous driving which requires real-time inference speed and high performance. Many works try to compromise spatial resolution to achieve real-time inference speed, which leads to poor performance. As a result, real-time segmentation task for embedded devices is still an open problem. In this paper, we focus on building a network with better performance possiblewhile still achieve real-time inference speed. We first use a pyramid kernel size to capture more spatial information instead of using just a 3×3 kernel size for DWConvolution in ShuffleNet v2. Meanwhile, an efficient Vortex Pooling module is employed to aggregate the contextual information and generate highresolution features. Compared with other state-of-the-art realtime semantic segmentation networks, the proposed network achieves similar inference speed and better performance on embedded device. Specifically, we achieve state-of-the-art 73.46{\%} mean IoU on Cityscapes test dataset, for a 768×1024 input, a speed of 46.1 frames per second on NVIDIA Jetson AGX Xavier embedded development board is achieved.",
keywords = "semantic segmentation, real-time, embedded",
author = "Jiansheng Dong and Jingling Yuan and Lin Li and Xian Zhong and Weiru Liu",
year = "2019",
month = "8",
day = "27",
language = "English",
booktitle = "31st International Conference on Tools with Artificial Intelligence (ICTAI2019)",
publisher = "Institute of Electrical and Electronics Engineers (IEEE)",
address = "United States",

}

RIS - suitable for import to EndNote

TY - GEN

T1 - An Efficient Semantic Segmentation Method using Pyramid ShuffleNet V2 with Vortex Pooling

AU - Dong, Jiansheng

AU - Yuan, Jingling

AU - Li, Lin

AU - Zhong, Xian

AU - Liu, Weiru

PY - 2019/8/27

Y1 - 2019/8/27

N2 - —Efficient and accurate semantic segmentation is particularly important especially for applications like autonomous driving which requires real-time inference speed and high performance. Many works try to compromise spatial resolution to achieve real-time inference speed, which leads to poor performance. As a result, real-time segmentation task for embedded devices is still an open problem. In this paper, we focus on building a network with better performance possiblewhile still achieve real-time inference speed. We first use a pyramid kernel size to capture more spatial information instead of using just a 3×3 kernel size for DWConvolution in ShuffleNet v2. Meanwhile, an efficient Vortex Pooling module is employed to aggregate the contextual information and generate highresolution features. Compared with other state-of-the-art realtime semantic segmentation networks, the proposed network achieves similar inference speed and better performance on embedded device. Specifically, we achieve state-of-the-art 73.46% mean IoU on Cityscapes test dataset, for a 768×1024 input, a speed of 46.1 frames per second on NVIDIA Jetson AGX Xavier embedded development board is achieved.

AB - —Efficient and accurate semantic segmentation is particularly important especially for applications like autonomous driving which requires real-time inference speed and high performance. Many works try to compromise spatial resolution to achieve real-time inference speed, which leads to poor performance. As a result, real-time segmentation task for embedded devices is still an open problem. In this paper, we focus on building a network with better performance possiblewhile still achieve real-time inference speed. We first use a pyramid kernel size to capture more spatial information instead of using just a 3×3 kernel size for DWConvolution in ShuffleNet v2. Meanwhile, an efficient Vortex Pooling module is employed to aggregate the contextual information and generate highresolution features. Compared with other state-of-the-art realtime semantic segmentation networks, the proposed network achieves similar inference speed and better performance on embedded device. Specifically, we achieve state-of-the-art 73.46% mean IoU on Cityscapes test dataset, for a 768×1024 input, a speed of 46.1 frames per second on NVIDIA Jetson AGX Xavier embedded development board is achieved.

KW - semantic segmentation

KW - real-time

KW - embedded

M3 - Conference contribution

BT - 31st International Conference on Tools with Artificial Intelligence (ICTAI2019)

PB - Institute of Electrical and Electronics Engineers (IEEE)

ER -