arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1805.04554cs.CV
收藏

ContextNet:实时的探索上下文和细节的语义分割

ContextNet: Exploring Context and Detail for Semantic Segmentation in Real-time

Toshiba Research(东芝研究院)

Rudra P K Poudel, Ujwal Bonde, Stephan Liwicki, Christopher Zach

摘要:

现代深度学习架构在许多具有挑战性的语义分割数据集上产生高度准确的结果。然而,现有技术的方法不能直接转移到实时的【形】应用或嵌入式设备,因为这种系统的天真适应以降低计算成本(速度、存储器和能量)导致准确度的显著下降。我们提出了ContextNet,这是一种新的深度神经网络架构,它建立在因子分解卷积,网络压缩和金字塔表示的基础上,以实时的【形】低内存需求产生有竞争力的语义分割。ContextNet将低分辨率的深度网络分支与关注高分辨率分割细节的浅分支相结合,前者有效地捕获全局上下文信息。我们在一项彻底的消融研究中分析了我们的网络,并在Cityscapes数据集上展示了结果,在全(1024x2048)分辨率下以每秒18.3帧的速度实现了66.1%的准确率(流数据的流水线计算为41.9 fps)。

英文摘要:

Modern deep learning architectures produce highly accurate results on many challenging semantic segmentation datasets. State-of-the-art methods are, however, not directly transferable to real-time applications or embedded devices, since naive adaptation of such systems to reduce computational cost (speed, memory and energy) causes a significant drop in accuracy. We propose ContextNet, a new deep neural network architecture which builds on factorized convolution, network compression and pyramid representation to produce competitive semantic segmentation in real-time with low memory requirement. ContextNet combines a deep network branch at low resolution that captures global context information efficiently with a shallow branch that focuses on high-resolution segmentation details. We analyse our network in a thorough ablation study and present results on the Cityscapes dataset, achieving 66.1% accuracy at 18.3 frames per second at full (1024x2048) resolution (41.9 fps with pipelined computations for streamed data).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑