iELAS: An ELAS-Based Energy-Efficient Accelerator for Real-Time Stereo Matching on FPGA Platform
iELAS: An ELAS-Based Energy-Efficient Accelerator for Real-Time Stereo Matching on FPGA Platform
复制标题
iELAS：基于 ELAS 的节能加速器，用于 FPGA 平台上的实时立体匹配
DOI：
10.1109/aicas51828.2021.9458401
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
A. Raychowdhury
中科院分区：
文献类型：
--
作者：
Tianren Gao;Zishen Wan;Yuyang Zhang;Bo Yu;Yanjun Zhang;Shaoshan Liu;A. Raychowdhury
立体声匹配是机器人导航和自动驾驶汽车的关键任务，提供了周围环境的深度估计。在所有立体声匹配算法中，有效的大规模立体声（ELAS）提供了效率和准确性之间最好的权衡之一。但是，由于固有的迭代过程和不可预测的内存访问模式，ELA只能在高端CPU上以1.5-3 fps运行，并且难以在低功率平台上实现实时性能。在本文中，我们为FPGA平台上的实时ELAS立体声匹配提出了一种节能架构。此外，原始的计算密集型和不规则的三角剖分模块是定期改革的，可以通过插值来进行插值，这更适合硬件。包括内存管理，并行性和管道上的优化，进一步用于减少内存足迹和改善吞吐量。与Intel I7 CPU和最先进的$ \ Mathrm {C} \ Mathrm {P} \ Mathrm {U}+$ FPGA实现相比，我们的FPGA实现可达到高达$ 38.4 \ timper $ $ 3.32 $和$ 3.32 \ Times $框架汇率提高，最高$ 27.1 \ times $和$ 1.13 \ times $ $ $提高能源效率。
Stereo matching is a critical task for robot navigation and autonomous vehicles, providing the depth estimation of surroundings. Among all stereo matching algorithms, Efficient Large-scale Stereo (ELAS) offers one of the best tradeoffs between efficiency and accuracy. However, due to the inherent iterative process and unpredictable memory access pattern, ELAS can only run at 1.5-3 fps on high-end CPUs and difficult to achieve real-time performance on low-power platforms. In this paper, we propose an energy-efficient architecture for real-time ELAS-based stereo matching on FPGA platform. Moreover, the original computational-intensive and irregular triangulation module is reformed in a regular manner with points interpolation, which is much more hardware-friendly. optimizations, including memory management, parallelism, and pipelining, are further utilized to reduce memory footprint and improve throughput. Compared with Intel i7 CPU and the state-of-the-art $\mathrm{C}\mathrm{P}\mathrm{U}+$FPGA implementation, our FPGA realization achieves up to $ 38.4\times$ and $ 3.32\times$ frame rate improvement, and up to $ 27.1\times$ and $ 1.13\times$ energy efficiency improvement, respectively.
影响因子：
9.2
作者：
Geiger, A.;Lenz, P.;Urtasun, R.
通讯作者：
Urtasun, R.