Next Article in Journal
Prototyping of an All-pMOS-Based Cross-Coupled Voltage Multiplier in Single-Well CMOS Technology for Energy Harvesting Utilizing a Gastric Acid Battery
Previous Article in Journal
A n-out-of-n Sharing Digital Image Scheme by Using Color Palette
Previous Article in Special Issue
A Review of Binarized Neural Networks
Article Menu
Issue 7 (July) cover image

Export Article

Open AccessArticle

Efficient Implementation of 2D and 3D Sparse Deconvolutional Neural Networks with a Uniform Architecture on FPGAs

College of Computer, National University of Defense Technology, Changsha 410073, China
*
Author to whom correspondence should be addressed.
Electronics 2019, 8(7), 803; https://doi.org/10.3390/electronics8070803
Received: 29 May 2019 / Revised: 6 July 2019 / Accepted: 9 July 2019 / Published: 18 July 2019
  |  
PDF [1194 KB, uploaded 18 July 2019]
  |  

Abstract

Three-dimensional (3D) deconvolution is widely used in many computer vision applications. However, most previous works have only focused on accelerating two-dimensional (2D) deconvolutional neural networks (DCNNs) on Field-Programmable Gate Arrays (FPGAs), while the acceleration of 3D DCNNs has not been well studied in depth as they have higher computational complexity and sparsity than 2D DCNNs. In this paper, we focus on the acceleration of both 2D and 3D sparse DCNNs on FPGAs by proposing efficient schemes for mapping 2D and 3D sparse DCNNs on a uniform architecture. Firstly, a pruning method is used to prune unimportant network connections and increase the sparsity of weights. After being pruned, the number of parameters of DCNNs is reduced significantly without accuracy loss. Secondly, the remaining non-zero weights are encoded in coordinate (COO) format, reducing the memory demands of parameters. Finally, to demonstrate the effectiveness of our work, we implement our accelerator design on the Xilinx VC709 evaluation platform for four real-life 2D and 3D DCNNs. After the first two steps, the storage required of DCNNs is reduced up to 3.9×. Results show that the performance of our method on the accelerator outperforms that of the our prior work by 2.5× to 3.6× in latency. View Full-Text
Keywords: DCNN; FPGA; pruning; sparsity; acceleration; 2D; 3D; uniform architecture DCNN; FPGA; pruning; sparsity; acceleration; 2D; 3D; uniform architecture
Figures

Figure 1

This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
SciFeed

Share & Cite This Article

MDPI and ACS Style

Wang, D.; Shen, J.; Wen, M.; Zhang, C. Efficient Implementation of 2D and 3D Sparse Deconvolutional Neural Networks with a Uniform Architecture on FPGAs. Electronics 2019, 8, 803.

Show more citation formats Show less citations formats

Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Related Articles

Article Metrics

Article Access Statistics

1

Comments

[Return to top]
Electronics EISSN 2079-9292 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top