Presentation | 2022-11-30 Deep Learning-based Hierarchical Object Detection System for High-Resolution Images Yusei Horikawa, Makoto Sugaya, Renpei Yoshida, Kazuma Mashiko, Tetsuya Matsumura, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper describes a new deep learning-based hierarchical object detection algorithm for high-resolution vision sensors. The proposed algorithm features hierarchical three-layers structure based on YOLO network. Wide area detection using a reduced image and local area detection using the original image are performed hierarchically for Full-HD images. These structure is anticipated to improve detecting performance as detect various size of objects. In evaluations with the VisDrone data set which is consisted by high-resolution images, our algorithm achieves about two times the number of detections and an improved mean average precision(mAP) of about 3% compared to the conventional algorithm. We also implemented this system on a small, low-cost JetsonNano board and confirmed real-time operation at approximately 3 frames per second. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Object Detection / High-Resolution Images / Deep Learning / YOLO / Embedded System |
Paper # | VLD2022-44,ICD2022-61,DC2022-60,RECONF2022-67 |
Date of Issue | 2022-11-21 (VLD, ICD, DC, RECONF) |
Conference Information | |
Committee | VLD / DC / RECONF / ICD / IPSJ-SLDM |
---|---|
Conference Date | 2022/11/28(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Kanazawa Bunka Hall |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Design Gaia 2022 -New Field of VLSI Design- |
Chair | Minako Ikeda(NTT) / Tatsuhiro Tsuchiya(Osaka Univ.) / Kentaro Sano(RIKEN) / Masafumi Takahashi(Kioxia) / Hiroyuki Ochi(Ritsumeikan Univ.) |
Vice Chair | Shigetoshi Nakatake(Univ. of Kitakyushu) / Toshinori Hosokawa(Nihon Univ.) / Yoshiki Yamaguchi(Tsukuba Univ.) / Tomonori Izumi(Ritsumeikan Univ.) / Makoto Ikeda(Univ. of Tokyo) |
Secretary | Shigetoshi Nakatake(NBS) / Toshinori Hosokawa(Hirosaki Univ.) / Yoshiki Yamaguchi(Nihon Univ.) / Tomonori Izumi(Chiba Univ.) / Makoto Ikeda(NEC) / (Toyohashi Univ. of Tech.) |
Assistant | Takuma Nishimoto(Hitachi) / / Yukitaka Takemura(INTEL) / Yasunori Osana(Ryukyu Univ.) / Yoshiaki Yoshihara(KIOXIA) / Jun Shiomi(Osaka Univ.) / Takeshi Kuboki(Sony Semiconductor Solutions) |
Paper Information | |
Registration To | Technical Committee on VLSI Design Technologies / Technical Committee on Dependable Computing / Technical Committee on Reconfigurable Systems / Technical Committee on Integrated Circuits and Devices / Special Interest Group on System and LSI Design Methodology |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Deep Learning-based Hierarchical Object Detection System for High-Resolution Images |
Sub Title (in English) | |
Keyword(1) | Object Detection |
Keyword(2) | High-Resolution Images |
Keyword(3) | Deep Learning |
Keyword(4) | YOLO |
Keyword(5) | Embedded System |
1st Author's Name | Yusei Horikawa |
1st Author's Affiliation | Nihon University(Nihon Univ.) |
2nd Author's Name | Makoto Sugaya |
2nd Author's Affiliation | Nihon University(Nihon Univ.) |
3rd Author's Name | Renpei Yoshida |
3rd Author's Affiliation | Nihon University(Nihon Univ.) |
4th Author's Name | Kazuma Mashiko |
4th Author's Affiliation | Nihon University(Nihon Univ.) |
5th Author's Name | Tetsuya Matsumura |
5th Author's Affiliation | Nihon University(Nihon Univ.) |
Date | 2022-11-30 |
Paper # | VLD2022-44,ICD2022-61,DC2022-60,RECONF2022-67 |
Volume (vol) | vol.122 |
Number (no) | VLD-283,ICD-284,DC-285,RECONF-286 |
Page | pp.pp.144-149(VLD), pp.144-149(ICD), pp.144-149(DC), pp.144-149(RECONF), |
#Pages | 6 |
Date of Issue | 2022-11-21 (VLD, ICD, DC, RECONF) |