激光与光电子学进展, 2020, 57 (2): 021017, 网络出版: 2020-01-03
基于膨胀卷积平滑及轻型上采样的实时语义分割 下载: 1374次
Real-Time Semantic Segmentation Based on Dilated Convolution Smoothing and Lightweight Up-Sampling
图像处理 实时语义分割 轻量级网络 知识蒸馏 膨胀卷积 轻型上采样 image processing real-time semantic segmentation lightweight network knowledge distillation dilated convolution lightweight up-sampling
摘要
针对轻量级网络在语义分割速度上较快,但精度较低的问题,在轻量级网络基础上,提出了基于膨胀卷积平滑及轻型上采样的实时语义分割方法。为了提高分割速度,采用结合结构化知识蒸馏的轻量级网络ResNeXt-18作为特征提取网络。设计了膨胀卷积平滑模块及轻型上采样模块,用来提高语义分割的精确度。为验证所提方法的有效性,在Cityscapes数据集及CamVid数据集上进行评估,在Cityscapes数据集上得到了速度为40.2 frame/s,精度为76.8%,参数量仅为1.18×10 7的结果。实验表明,本文提出的实时语义分割方法在保持方法实时性的同时可以得到较好的分割准确度,具有一定的实用价值。
Abstract
In lightweight networks, the speed of semantic segmentation is high but the accuracy is low. On the basis of lightweight networks, a real-time semantic segmentation method based on dilated convolution smoothing and lightweight up-sampling is proposed. To improve segmentation speed, a lightweight network, ResNeXt-18, with structured knowledge distillation is used as feature extraction network. To improve the segmentation accuracy, a dilated convolution smoothing module and a lightweight up-sampling module are designed. To verify the effectiveness of the proposed method, the evaluations are carried out using the Cityscapes and CamVid datasets, obtaining the speed of 40.2 frame/s and the segmentation accuracy of 76.8%, with a parameter count of 1.18×10 7. The experimental results demonstrate that the proposed method can obtain high segmentation accuracy while maintaining its high-speed real-time performance; as such, it has certain practical value.
程晓悦, 赵龙章, 胡穹, 史家鹏. 基于膨胀卷积平滑及轻型上采样的实时语义分割[J]. 激光与光电子学进展, 2020, 57(2): 021017. 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): 021017.