arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1708.02550cs.CVcs.RO
收藏

面向自动驾驶的快速场景理解

Fast Scene Understanding for Autonomous Driving

ESAT-PSI, KU Leuven(鲁汶大学 ESAT-PSI)

Davy Neven, Bert De Brabandere, Stamatios Georgoulis, Marc Proesmans, Luc Van Gool

摘要:

大多数实例感知语义标记方法传统上关注准确性。运行时间和内存占用等其他方面对于自动驾驶等实时应用也同样重要。出于这种观察和最近的作品,解决多个任务与一个单一的集成架构的启发,在本文中,我们提出了一个实时有效的实现的基础上ENet,解决了三个自动驾驶相关的任务:语义场景分割,实例分割和单眼深度估计。我们的方法建立在一个分支的ENet架构与共享的编码器,但不同的解码器分支的三个任务。与单独运行每个任务相比,所提出的方法可以在Cityscapes数据集上以1024x512的分辨率以21 fps运行,而不会牺牲准确性。

英文摘要:

Most approaches for instance-aware semantic labeling traditionally focus on accuracy. Other aspects like runtime and memory footprint are arguably as important for real-time applications such as autonomous driving. Motivated by this observation and inspired by recent works that tackle multiple tasks with a single integrated architecture, in this paper we present a real-time efficient implementation based on ENet that solves three autonomous driving related tasks at once: semantic scene segmentation, instance segmentation and monocular depth estimation. Our approach builds upon a branched ENet architecture with a shared encoder but different decoder branches for each of the three tasks. The presented method can run at 21 fps at a resolution of 1024x512 on the Cityscapes dataset without sacrificing accuracy compared to running each task separately.

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。

↑