Committee |
Date Time |
Place |
Paper Title / Authors |
Abstract |
Paper # |
PRMU, IBISML, IPSJ-CVIM |
2024-03-03 17:00 |
Hiroshima |
Hiroshima Univ. Higashi-Hiroshima campus (Primary: On-site, Secondary: Online) |
Multi-agent reinforcement learning based control method for large-scale crowd movement on Mojiko Fireworks Festival dataset Kazuya Miyazaki, Masato Kiyama, Motoki Amagasaki, Toshiaki Okamoto (Kumamoto Univ.) IBISML2023-45 |
The importance of human flow guidance is increasing in response to accidents at events. In recent years, some research h... [more] |
IBISML2023-45 pp.36-43 |
IBISML |
2023-12-21 10:30 |
Tokyo |
National Institute of Informatics (Primary: On-site, Secondary: Online) |
Badminton Rally Analysis Using LSTM Atsushi Yoshinaga, Masato Kiyama, Motoki Amagasaki (Kumamoto Univ.) IBISML2023-35 |
In this study, we analyze badminton rallies as a tactical support technology for sports using AI. In badminton, it is re... [more] |
IBISML2023-35 pp.31-36 |
RECONF |
2022-09-08 10:10 |
Aichi |
emCAMPUS STUDIO (Primary: On-site, Secondary: Online) |
Proposal and evaluation of Combined Posit MAC unit (CPMAC) for both DNN inference and training Yuta Masuda, Yasuhiro Nakahara, Masato Kiyama, Masahiro Iida (Kumamoto Univ.) RECONF2022-34 |
Recently, there has been a lot of research on DNN hardware accelerators for the edge that use Posit as a number represen... [more] |
RECONF2022-34 pp.29-34 |
NC, IBISML, IPSJ-BIO, IPSJ-MPS [detail] |
2022-06-29 13:30 |
Okinawa |
(Primary: On-site, Secondary: Online) |
Acceleration of data assimilation for Large-scale human flow data Miyazaki Kazuya, Kiyama Masato, Amagasaki Motoki, Okamoto Toshiaki (Kumamoto Univ) NC2022-25 IBISML2022-25 |
Data assimilation has been attracting attention as the importance of understanding human flow has been emphasized in the... [more] |
NC2022-25 IBISML2022-25 pp.178-183 |
VLD, DC, RECONF, ICD, IPSJ-SLDM (Joint) [detail] |
2021-12-01 10:35 |
Online |
Online |
Basic evaluation of ReNA, a DNN accelerator using numerical representation posit Yasuhiro Nakahara, Yuta Masuda, Masato Kiyama, Motoki Amagasaki, Masahiro Iida (Kumamoto Univ.) VLD2021-24 ICD2021-34 DC2021-30 RECONF2021-32 |
In Convolutional Neural Network (CNN) accelerators for edge, numerical precision of data should be reduced as much as po... [more] |
VLD2021-24 ICD2021-34 DC2021-30 RECONF2021-32 pp.43-48 |
HWS, VLD [detail] |
2021-03-03 11:15 |
Online |
Online |
The Design and Development of of Quantized Neural Networks Library for Exact Hardware Emulation Masato Kiyama, Yasuhiro Nakahara, Motoki Amagasaki, Masahiro Iida (Kumamoto Univ.) VLD2020-70 HWS2020-45 |
Quantization is used to speed up execution time and save power when runnning Deep neural networks (DNNs) on edge devices... [more] |
VLD2020-70 HWS2020-45 pp.18-23 |
NC, NLP (Joint) |
2021-01-21 12:05 |
Online |
Online |
Examination of precipitation estimation using atmospheric variables Takanori Ito, Motoki Amagasaki, Kei Ishida, Masato Kiyama, Masahiro Iida (GSST Kumamoto University) NC2020-34 |
In this paper, we developed a model for SR using ConvLSTM to improve the resolution of precipitation data.
In the relat... [more] |
NC2020-34 pp.13-17 |
MBE, NC (Joint) |
2020-12-18 14:50 |
Online |
Online |
Super resolution for sea surface temperature with CNN and GAN Tomoki Izumi, Motoki Amagasaki, Kei Ishida, Masato Kiyama (Kumamoto Univ.) NC2020-28 |
In this paper, we use the deep neural networks (DNN)-based single image super-resolution (SISR) method for the super res... [more] |
NC2020-28 pp.1-6 |
VLD, DC, RECONF, ICD, IPSJ-SLDM (Joint) [detail] |
2020-11-17 10:45 |
Online |
Online |
Implementation of YOLO in the AI accelerator ReNA Toma Uemura, Yasuhiro Nakahara, Motoki Amagasaki, Masato Kiyama, Masahiro Iida (Kumamoto Univ.) VLD2020-22 ICD2020-42 DC2020-42 RECONF2020-41 |
The object detection,which is a typical AI process,has been attracting attention in various fields because it can identi... [more] |
VLD2020-22 ICD2020-42 DC2020-42 RECONF2020-41 pp.66-71 |
RECONF |
2019-09-20 14:00 |
Fukuoka |
KITAKYUSHU Convention Center |
Quantized Neural Networks Library for Exact Hardware Emulation Masato Kiyama, Yasuhiro Nakahara, Motoki Amagasaki, Masahiro Iida (Kumamoto Univ.) RECONF2019-33 |
Deep neural networks (DNNs) have recently shown outstanding performance in many application domains.
However, it is dif... [more] |
RECONF2019-33 pp.69-74 |
RECONF |
2019-05-10 13:55 |
Tokyo |
Tokyo Tech Front |
Deep Learning Framework with Numerical Precision Masato Kiyama, Motoki Amagasaki, Masahiro Iida (Kumamoto Univ.) RECONF2019-15 |
[more] |
RECONF2019-15 pp.79-84 |
WIT |
2008-03-23 14:45 |
Fukuoka |
Kitakyushu Science and Research Park |
Proposal of a Weighted-SPEAK method for cochlear implant system and design of its evaluation simulator Masayuki Satou (Kumamoto PCT), Masato Ikiyama, Dan Nishikido, Hiroki Toyoshima, Tadashi Sakata, Yuichi Ueda (Kumamoto Univ.) WIT2007-105 |
A hybrid method, which have proposed as a speech processor for the cochlear implant, is a mixture type of a general-purp... [more] |
WIT2007-105 pp.85-90 |