Object Detection Inference System with Deep Learning NVDLA Accelerator

Yeong Kang Lai; Chuan Wei Huang

doi:10.1109/ICCE-Taiwan55306.2022.9869085

Object Detection Inference System with Deep Learning NVDLA Accelerator

Yeong Kang Lai, Chuan Wei Huang

National Chung Hsing University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Nowadays, when Artificial Intelligence (AI) is getting more advanced, many applications are following its footsteps. For example, in the fields of object recognition and speech processing, there are many different DNN (deep neural network) structures. By combining these DNNs and collecting a large library of images or audio. It uses GPUs for training to achieve high accuracy and meet its application requirements. This paper focuses on object recognition in store applications and uses DNN hardware accelerators for real-time inference. It maps to NVDLA hardware using YOLOv2-tiny. NVDLA can perform YOLOv2-tiny with 256 MAC@150 MHz frequency. Also, it can reach up to 5 FPS.

Original language	English
Title of host publication	Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	67-68
Number of pages	2
ISBN (Electronic)	9781665470506
DOIs	https://doi.org/10.1109/ICCE-Taiwan55306.2022.9869085
State	Published - 2022
Externally published	Yes
Event	2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022 - Taipei, Taiwan Duration: 06 07 2022 → 08 07 2022

Publication series

Name	Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022

Conference

Conference	2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
Country/Territory	Taiwan
City	Taipei
Period	06/07/22 → 08/07/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/ICCE-Taiwan55306.2022.9869085

Cite this

Lai, Y. K., & Huang, C. W. (2022). Object Detection Inference System with Deep Learning NVDLA Accelerator. In Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022 (pp. 67-68). (Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCE-Taiwan55306.2022.9869085

Lai, Yeong Kang ; Huang, Chuan Wei. / Object Detection Inference System with Deep Learning NVDLA Accelerator. Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 67-68 (Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022).

@inproceedings{1e567e491fed4bde805f99edeca56c74,

title = "Object Detection Inference System with Deep Learning NVDLA Accelerator",

abstract = "Nowadays, when Artificial Intelligence (AI) is getting more advanced, many applications are following its footsteps. For example, in the fields of object recognition and speech processing, there are many different DNN (deep neural network) structures. By combining these DNNs and collecting a large library of images or audio. It uses GPUs for training to achieve high accuracy and meet its application requirements. This paper focuses on object recognition in store applications and uses DNN hardware accelerators for real-time inference. It maps to NVDLA hardware using YOLOv2-tiny. NVDLA can perform YOLOv2-tiny with 256 MAC@150 MHz frequency. Also, it can reach up to 5 FPS.",

author = "Lai, {Yeong Kang} and Huang, {Chuan Wei}",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022 ; Conference date: 06-07-2022 Through 08-07-2022",

year = "2022",

doi = "10.1109/ICCE-Taiwan55306.2022.9869085",

language = "英语",

series = "Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "67--68",

booktitle = "Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022",

address = "美国",

}

Lai, YK & Huang, CW 2022, Object Detection Inference System with Deep Learning NVDLA Accelerator. in Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022. Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022, Institute of Electrical and Electronics Engineers Inc., pp. 67-68, 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022, Taipei, Taiwan, 06/07/22. https://doi.org/10.1109/ICCE-Taiwan55306.2022.9869085

Object Detection Inference System with Deep Learning NVDLA Accelerator. / Lai, Yeong Kang; Huang, Chuan Wei.
Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 67-68 (Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Object Detection Inference System with Deep Learning NVDLA Accelerator

AU - Lai, Yeong Kang

AU - Huang, Chuan Wei

PY - 2022

Y1 - 2022

N2 - Nowadays, when Artificial Intelligence (AI) is getting more advanced, many applications are following its footsteps. For example, in the fields of object recognition and speech processing, there are many different DNN (deep neural network) structures. By combining these DNNs and collecting a large library of images or audio. It uses GPUs for training to achieve high accuracy and meet its application requirements. This paper focuses on object recognition in store applications and uses DNN hardware accelerators for real-time inference. It maps to NVDLA hardware using YOLOv2-tiny. NVDLA can perform YOLOv2-tiny with 256 MAC@150 MHz frequency. Also, it can reach up to 5 FPS.

AB - Nowadays, when Artificial Intelligence (AI) is getting more advanced, many applications are following its footsteps. For example, in the fields of object recognition and speech processing, there are many different DNN (deep neural network) structures. By combining these DNNs and collecting a large library of images or audio. It uses GPUs for training to achieve high accuracy and meet its application requirements. This paper focuses on object recognition in store applications and uses DNN hardware accelerators for real-time inference. It maps to NVDLA hardware using YOLOv2-tiny. NVDLA can perform YOLOv2-tiny with 256 MAC@150 MHz frequency. Also, it can reach up to 5 FPS.

UR - http://www.scopus.com/inward/record.url?scp=85138712768&partnerID=8YFLogxK

U2 - 10.1109/ICCE-Taiwan55306.2022.9869085

DO - 10.1109/ICCE-Taiwan55306.2022.9869085

M3 - 会议稿件

AN - SCOPUS:85138712768

T3 - Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022

SP - 67

EP - 68

BT - Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022

Y2 - 6 July 2022 through 8 July 2022

ER -

Lai YK, Huang CW. Object Detection Inference System with Deep Learning NVDLA Accelerator. In Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 67-68. (Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022). doi: 10.1109/ICCE-Taiwan55306.2022.9869085

Object Detection Inference System with Deep Learning NVDLA Accelerator

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this