Faculty Profiles - FUJII Keisuke

Papers - FUJII Keisuke

Division display 121 - 130 of about 130 ／ All the affair displays >>

協調に関する議論に向けたアプローチの提案－集団運動からみる他者の行動予測と適応－ Reviewed Open Access

市川淳, 藤井慶輔

認知科学 Vol. 27 ( 3 ) page： xx - xx 2020

Succinct Trie Index for Fast and Memory-Efficient Trajectory Similarity Search Reviewed Open Access

KANDA Shunsuke, TAKEUCHI Koh, FUJII Keisuke, TABEI Yasuo

Proceedings of the Annual Conference of JSAI Vol. 2020 ( 0 ) page： 2C4OS7a01 - 2C4OS7a01 2020

Publishing type：Research paper (scientific journal) Publisher：The Japanese Society for Artificial Intelligence

<p>This paper proposes an efficient similarity search method for massive collections of trajectories. This method enables fast similarity searches by leveraging locality sensitive hashing and tries. To achieve memory efficiency, we propose two novel techniques of node reduction and a space-efficient representation for tries. We experimentally test our method on its ability to retrieve similar trajectories for a query from large collections of trajectories and show that our method performs superiorly with respect to search time and memory efficiency.</p>

DOI： 10.11517/pjsai.JSAI2020.0_2C4OS7a01

Open Access

CiNii Research

「現場と研究の橋渡し」（ 2 ）にあたって Open Access

藤井慶輔, 永田直也

バスケットボール研究 Vol. 6 ( 0 ) page： 1 - 2 2020

Dynamic mode decomposition via dictionary learning for foreground modeling in videos Reviewed

Israr Ul Haq, Keisuke Fujii, Yoshinobu Kawahara

VISIGRAPP 2020 - Proceedings of the 15th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications Vol. 5 page： 476 - 483 2020

Language：English Publishing type：Research paper (international conference proceedings) Publisher：SciTePress

Accurate extraction of foregrounds in videos is one of the challenging problems in computer vision. In this study, we propose dynamic mode decomposition via dictionary learning (dl-DMD), which is applied to extract moving objects by separating the sequence of video frames into foreground and background information with a dictionary learned using block patches on the video frames. Dynamic mode decomposition (DMD) decomposes spatiotemporal data into spatial modes, each of whose temporal behavior is characterized by a single frequency and growth/decay rate and is applicable to split a video into foregrounds and the background when applying it to a video. And, in dl-DMD, DMD is applied on coefficient matrices estimated over a learned dictionary, which enables accurate estimation of dynamical information in videos. Due to this scheme, dl-DMD can analyze the dynamics of respective regions in a video based on estimated amplitudes and temporal evolution over patches. The results on synthetic data exhibit that dl-DMD outperforms the standard DMD and compressed DMD (cDMD) based methods. Also, the results of an empirical performance evaluation in the case of foreground extraction from videos using publicly available dataset demonstrates the effectiveness of the proposed dl-DMD algorithm and achieves a performance that is comparable to that of the state-of-the-art techniques in foreground extraction tasks.

DOI： 10.5220/0009144604760483

Scopus

researchmap

Motor control of practice and actual strokes by professional and amateur golfers differ but feature a distance-dependent control strategy. Reviewed International journal

Yumiko Hasegawa, Keisuke Fujii, Akito Miura, Keiko Yokoyama, Yuji Yamamoto

European journal of sport science Vol. 19 ( 9 ) page： 1204 - 1213 2019.10

Language：English Publishing type：Research paper (scientific journal)

We explored how practice and actual putting strokes differed between professionals and high-level golf amateurs, and how practice strokes reflected subtle differences in putting distances. We analysed swing amplitude, impact velocity, and acceleration profile of the club-head. The acceleration profiles showed that the motor control pattern of the practice stroke differed from that of the actual stroke. To clarify the effects of different putting distances on the practice stroke and to analyse how much the actual stroke could be explained by the practice stroke, we conducted individual regression analyses. The practice strokes of all participants could be divided into three strategies and five types by the coefficient of determination and the slope. This implies that the purpose of the practice stroke varied among golfers. Most golfers used the individual velocity criteria in their practice strokes, which resulted in different putting distances based on their criteria. Unexpectedly, we found no significant difference in skill level between professionals and high-level amateurs. The results of this study imply that the practice stroke does not duplicate the actual stroke, even for professional golfers with excellent skills. However, most high-level golfers adopted distance-dependent control strategies for slightly different putting distances.

DOI： 10.1080/17461391.2019.1595159

The kinetic mechanisms of vertical pointing movements. Reviewed International journal Open Access

Shinji Yamamoto, Keisuke Fujii, Kisho Zippo, Keisuke Kushiro, Masanobu Araki

Heliyon Vol. 5 ( 7 ) page： e02012 2019.7

Language：English Publishing type：Research paper (scientific journal)

The present study utilized induced acceleration analysis to clarify the contributions of muscular and gravitational torques to the kinematics of vertical pointing movements performed by the upper limb. The study included eight healthy men with a mean age of 25 years. The experiment was divided into three blocks with ten trials in each, comprising five upward and five downward, randomly executed movements. The movements were recorded by a motion capture system and were subsequently analyzed. During the deceleration phase of the upward movement and the acceleration phase of the downward movement, the angular acceleration induced by gravitational torque contributed more to the generation of net induced angular acceleration than the angular acceleration induced by muscular torque. In addition, the difference between the net induced angular acceleration profiles during the upward and downward movements was mainly attributable to the difference between the respective angular acceleration profiles induced by muscular torque. These findings suggest that the central nervous system considers the gravitational effect on the upper limb in a phase-specific manner and accordingly generates a torque-derived kinematic difference with respect to the movement direction.

DOI： 10.1016/j.heliyon.2019.e02012

Open Access

PubMed

researchmap

Analysis of factors predicting who obtains a ball in basketball rebounding situations Reviewed

Motokazu Hojo, Keisuke Fujii, Yoshinobu Kawahara

International Journal of Performance Analysis in Sport Vol. 19 ( 2 ) page： 192 - 205 2019.2

Supervised Dynamic Mode Decomposition via Multitask Learning Reviewed

Keisuke Fujii, Yoshinobu Kawahara

Pattern Recognition Letters Vol. 122 ( 1 ) page： 7 - 13 2019.2

指導者から見た「研究と現場の橋渡し」 Open Access

陸川章, 小山孟志, 青木美帆, 藤井慶輔

バスケットボール研究 Vol. 5 ( 0 ) page： 17 - 25 2019

“Bridging the gap between the field and the laboratory” Open Access

Fujii Keisuke, Nagata Naoya

The Japan Journal of Basketball Studies Vol. 5 ( 0 ) page： 1 - 2 2019

　 PREV - NEXT 　

<FUJII Keisuke>

Papers - FUJII Keisuke