HARFLOW3D: A Latency-Oriented 3D-CNN Accelerator Toolflow for HAR on FPGA Devices

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Toupas, Petros, Montgomerie-Corcoran, Alexander, Bouganis, Christos-Savvas, Tzovaras, Dimitrios
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929263282749440
author Toupas, Petros
Montgomerie-Corcoran, Alexander
Bouganis, Christos-Savvas
Tzovaras, Dimitrios
author_facet Toupas, Petros
Montgomerie-Corcoran, Alexander
Bouganis, Christos-Savvas
Tzovaras, Dimitrios
contents For Human Action Recognition tasks (HAR), 3D Convolutional Neural Networks have proven to be highly effective, achieving state-of-the-art results. This study introduces a novel streaming architecture based toolflow for mapping such models onto FPGAs considering the model's inherent characteristics and the features of the targeted FPGA device. The HARFLOW3D toolflow takes as input a 3D CNN in ONNX format and a description of the FPGA characteristics, generating a design that minimizes the latency of the computation. The toolflow is comprised of a number of parts, including i) a 3D CNN parser, ii) a performance and resource model, iii) a scheduling algorithm for executing 3D models on the generated hardware, iv) a resource-aware optimization engine tailored for 3D models, v) an automated mapping to synthesizable code for FPGAs. The ability of the toolflow to support a broad range of models and devices is shown through a number of experiments on various 3D CNN and FPGA system pairs. Furthermore, the toolflow has produced high-performing results for 3D CNN models that have not been mapped to FPGAs before, demonstrating the potential of FPGA-based systems in this space. Overall, HARFLOW3D has demonstrated its ability to deliver competitive latency compared to a range of state-of-the-art hand-tuned approaches being able to achieve up to 5$\times$ better performance compared to some of the existing works.
format Preprint
id arxiv_https___arxiv_org_abs_2303_17218
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle HARFLOW3D: A Latency-Oriented 3D-CNN Accelerator Toolflow for HAR on FPGA Devices
Toupas, Petros
Montgomerie-Corcoran, Alexander
Bouganis, Christos-Savvas
Tzovaras, Dimitrios
Hardware Architecture
Computer Vision and Pattern Recognition
Machine Learning
For Human Action Recognition tasks (HAR), 3D Convolutional Neural Networks have proven to be highly effective, achieving state-of-the-art results. This study introduces a novel streaming architecture based toolflow for mapping such models onto FPGAs considering the model's inherent characteristics and the features of the targeted FPGA device. The HARFLOW3D toolflow takes as input a 3D CNN in ONNX format and a description of the FPGA characteristics, generating a design that minimizes the latency of the computation. The toolflow is comprised of a number of parts, including i) a 3D CNN parser, ii) a performance and resource model, iii) a scheduling algorithm for executing 3D models on the generated hardware, iv) a resource-aware optimization engine tailored for 3D models, v) an automated mapping to synthesizable code for FPGAs. The ability of the toolflow to support a broad range of models and devices is shown through a number of experiments on various 3D CNN and FPGA system pairs. Furthermore, the toolflow has produced high-performing results for 3D CNN models that have not been mapped to FPGAs before, demonstrating the potential of FPGA-based systems in this space. Overall, HARFLOW3D has demonstrated its ability to deliver competitive latency compared to a range of state-of-the-art hand-tuned approaches being able to achieve up to 5$\times$ better performance compared to some of the existing works.
title HARFLOW3D: A Latency-Oriented 3D-CNN Accelerator Toolflow for HAR on FPGA Devices
topic Hardware Architecture
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2303.17218