Analysis of the acceleration of deep learning inference models on a heterogeneous architecture based on OpenVINO

Guerrouj, Fatima Zahra; ABOUZAHIR, Mohamed; Ramzi, Mustapha; ABDALI, El Mehdi

doi:10.1109/isaect53699.2021.9668607

Cited by 3 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…YOLOv4 GEMM operations are implemented on DE5a-Net DDR4, enabling highspeed data processing with high-performance Intel Arria 10 FPGAs. The board includes 16 GB of DDR4 memory running at over 75 Gbps, allowing data transfer up to 7.876 Gbps over PCIe Gen 3 × 8 between the board and the host PC [34]. In addition to these features, the host PC is equipped with an 8-core Intel Xeon Silver 4108 processor that runs at a basic frequency of 1.8 GHz.…”

Section: Resultsmentioning

confidence: 99%

Efficient GEMM Implementation for Vision-Based Object Detection in Autonomous Driving Applications

Guerrouj

Rodríguez

Abouzahir³

et al. 2023

JLPEA

Self Cite

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have been incredibly effective for object detection tasks. YOLOv4 is a state-of-the-art object detection algorithm designed for embedded systems. It is based on YOLOv3 and has improved accuracy, speed, and robustness. However, deploying CNNs on embedded systems such as Field Programmable Gate Arrays (FPGAs) is difficult due to their limited resources. To address this issue, FPGA-based CNN architectures have been developed to improve the resource utilization of CNNs, resulting in improved accuracy and speed. This paper examines the use of General Matrix Multiplication Operations (GEMM) to accelerate the execution of YOLOv4 on embedded systems. It reviews the most recent GEMM implementations and evaluates their accuracy and robustness. It also discusses the challenges of deploying YOLOv4 on autonomous vehicle datasets. Finally, the paper presents a case study demonstrating the successful implementation of YOLOv4 on an Intel Arria 10 embedded system using GEMM.

show abstract

Section: Resultsmentioning

confidence: 99%