DCSN: DEFORMABLE CONVOLUTIONAL SEMANTIC SEGMENTATION NEURAL NETWORK FOR NON-RIGID SCENES

Bor Sheng Huang, Chih Chung Hsu, Wo Ting Liao, Han Yi Kao, Xian Yun Wang

研究成果: Conference contribution

1 引文 斯高帕斯(Scopus)

摘要

This paper presents a novel semantic segmentation network for outdoor and unstructured scenarios for autonomous driving based on deformable convolution and geometric distortion pipelines. The semantic segmentation tasks for autonomous driving are generally designed for the urban scene, city-view, and highly structured scenarios, such as the CityScapes dataset, KITTI, and BDD, while rare study focuses on outskirts scenarios. Therefore, the performance of existing semantic segmentation networks on such datasets might be unreliable. To conquer this issue, a novel densely connected residual block (DCRB) with the deformable convolution is proposed to form our backbone for capturing the non-rigid feature representation. In this way, the gradient flow of our DCRB could be better back-propagated from the segmentation head, resulting in a stable training process. Second, geometric distortion augmentation is introduced in the data augmentation pipeline, simulating the possible deformation situations in real-world outdoor scenarios. The experiments are conducted that the proposed semantic segmentation network significantly outperforms the state-of-the-art methods for both Cityscapes and Outdoor scenarios.

原文English
主出版物標題2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
發行者Institute of Electrical and Electronics Engineers Inc.
頁面4668-4672
頁數5
ISBN(電子)9781665405409
DOIs
出版狀態Published - 2022
事件2022 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2022 - Hybrid, Singapore
持續時間: 2022 5月 222022 5月 27

出版系列

名字ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
2022-May
ISSN(列印)1520-6149

Conference

Conference2022 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2022
國家/地區Singapore
城市Hybrid
期間22-05-2222-05-27

All Science Journal Classification (ASJC) codes

  • 軟體
  • 訊號處理
  • 電氣與電子工程

指紋

深入研究「DCSN: DEFORMABLE CONVOLUTIONAL SEMANTIC SEGMENTATION NEURAL NETWORK FOR NON-RIGID SCENES」主題。共同形成了獨特的指紋。

引用此