Search

Your search keyword '"Sakaguchi, Taichi"' showing total 840 results
Start Over Author "Sakaguchi, Taichi"
840 results on '"Sakaguchi, Taichi"'

2. Object Instance Retrieval in Assistive Robotics: Leveraging Fine-Tuned SimSiam with Multi-View Images Based on 3D Semantic Map

Author

Sakaguchi, Taichi, Taniguchi, Akira, Hagiwara, Yoshinobu, Hafi, Lotfi El, Hasegawa, Shoichi, and Taniguchi, Tadahiro

Subjects
Computer Science - Robotics
Abstract

Robots that assist humans in their daily lives should be able to locate specific instances of objects in an environment that match a user's desired objects. This task is known as instance-specific image goal navigation (InstanceImageNav), which requires a model that can distinguish different instances of an object within the same class. A significant challenge in robotics is that when a robot observes the same object from various 3D viewpoints, its appearance may differ significantly, making it difficult to recognize and locate accurately. In this paper, we introduce a method called SimView, which leverages multi-view images based on a 3D semantic map of an environment and self-supervised learning using SimSiam to train an instance-identification model on-site. The effectiveness of our approach was validated using a photorealistic simulator, Habitat Matterport 3D, created by scanning actual home environments. Our results demonstrate a 1.7-fold improvement in task accuracy compared with contrastive language-image pre-training (CLIP), a pre-trained multimodal contrastive learning method for object searching. This improvement highlights the benefits of our proposed fine-tuning method in enhancing the performance of assistive robots in InstanceImageNav tasks. The project website is https://emergentsystemlabstudent.github.io/MultiViewRetrieve/., Comment: See website at https://emergentsystemlabstudent.github.io/MultiViewRetrieve/. Accepted to IROS2024

Published
2024

3. Real-world Instance-specific Image Goal Navigation: Bridging Domain Gaps via Contrastive Learning

Author

Sakaguchi, Taichi, Taniguchi, Akira, Hagiwara, Yoshinobu, Hafi, Lotfi El, Hasegawa, Shoichi, and Taniguchi, Tadahiro

Subjects
Computer Science - Robotics, Computer Science - Computation and Language, Computer Science - Computer Vision and Pattern Recognition
Abstract

Improving instance-specific image goal navigation (InstanceImageNav), which locates the identical object in a real-world environment from a query image, is essential for robotic systems to assist users in finding desired objects. The challenge lies in the domain gap between low-quality images observed by the moving robot, characterized by motion blur and low-resolution, and high-quality query images provided by the user. Such domain gaps could significantly reduce the task success rate but have not been the focus of previous work. To address this, we propose a novel method called Few-shot Cross-quality Instance-aware Adaptation (CrossIA), which employs contrastive learning with an instance classifier to align features between massive low- and few high-quality images. This approach effectively reduces the domain gap by bringing the latent representations of cross-quality images closer on an instance basis. Additionally, the system integrates an object image collection with a pre-trained deblurring model to enhance the observed image quality. Our method fine-tunes the SimSiam model, pre-trained on ImageNet, using CrossIA. We evaluated our method's effectiveness through an InstanceImageNav task with 20 different types of instances, where the robot identifies the same instance in a real-world environment as a high-quality query image. Our experiments showed that our method improves the task success rate by up to three times compared to the baseline, a conventional approach based on SuperGlue. These findings highlight the potential of leveraging contrastive learning and image enhancement techniques to bridge the domain gap and improve object localization in robotic applications. The project website is https://emergentsystemlabstudent.github.io/DomainBridgingNav/., Comment: See website at https://emergentsystemlabstudent.github.io/DomainBridgingNav/. Accepted to IEEE IRC2024

Published
2024

8. An Estimation Framework for Passerby Engagement Interacting with Social Robots

Author

Sakaguchi, Taichi, Okafuji, Yuki, Matsumura, Kohei, Baba, Jun, and Nakanishi, Junya

Subjects
Computer Science - Robotics
Abstract

Social robots are expected to be a human labor support technology, and one application of them is an advertising medium in public spaces. When social robots provide information, such as recommended shops, adaptive communication according to the user's state is desired. User engagement, which is also defined as the level of interest in the robot, is likely to play an important role in adaptive communication. Therefore, in this paper, we propose a new framework to estimate user engagement. The proposed method focuses on four unsolved open problems: multi-party interactions, process of state change in engagement, difficulty in annotating engagement, and interaction dataset in the real world. The accuracy of the proposed method for estimating engagement was evaluated using interaction duration. The results show that the interaction duration can be accurately estimated by considering the influence of the behaviors of other people; this also implies that the proposed model accurately estimates the level of engagement during interaction with the robot.

Published
2022

13. Impact of complete revascularization in coronary artery bypass grafting for ischemic cardiomyopathy

Author

Nakae, Masaro, Kainuma, Satoshi, Toda, Koichi, Yoshikawa, Yasushi, Hata, Hiroki, Yoshioka, Daisuke, Kawamura, Takuji, Kawamura, Ai, Kashiyama, Noriyuki, Ueno, Takayoshi, Kuratani, Toru, Kondoh, Haruhiko, Hiraoka, Arudo, Sakaguchi, Taichi, Yoshitaka, Hidenori, Shirakawa, Yukitoshi, Takahashi, Toshiki, Sakaki, Masayuki, Masai, Takafumi, Komukai, Sho, Kitamura, Tetsuhisa, Hirayama, Atsushi, Shimomura, Yoshimitsu, and Miyagawa, Shigeru

Published
2023
Full Text
View/download PDF

17. Surgical Ablation Concomitant With Nonmitral Valve Surgery for Persistent Atrial Fibrillation

Author

Kainuma, Satoshi, Mitsuno, Masataka, Toda, Koichi, Miyagawa, Shigeru, Yoshikawa, Yasushi, Hata, Hiroki, Yoshioka, Daisuke, Kawamura, Takuji, Kawamura, Ai, Kashiyama, Noriyuki, Ueno, Takayoshi, Kuratani, Toru, Kondoh, Haruhiko, Funatsu, Toshihiro, Hiraoka, Arudo, Sakaguchi, Taichi, Yoshitaka, Hidenori, Shirakawa, Yukitoshi, Takahashi, Toshiki, Sakaki, Masayuki, Masai, Takafumi, Saito, Shunsuke, Monta, Osamu, Kitamura, Tetsuhisa, Komukai, Sho, Hirayama, Atsushi, Taniguchi, Kazuhiro, Miyamoto, Yuji, and Sawa, Yoshiki

Published
2021
Full Text
View/download PDF

18. Multiple percutaneous coronary interventions worsen outcomes for subsequent surgical correction of chronic ischemic mitral regurgitation

Author

Yoshikawa, Yasushi, Hata, Hiroki, Funatsu, Toshihiro, Masai, Takafumi, Shirakawa, Yukitoshi, Takahashi, Toshiki, Nishi, Hiroyuki, Kawamura, Masashi, Monta, Osamu, Yamauchi, Takashi, Kainuma, Satoshi, Toda, Koichi, Miyagawa, Shigeru, Yoshioka, Daisuke, Kawamura, Takuji, Kawamura, Ai, Kashiyama, Noriyuki, Kuratani, Toru, Yokoi, Kensuke, Ide, Seiko, Mizote, Isamu, Kioka, Hidetaka, Ohtani, Tomohito, Hikoso, Shungo, Kondoh, Haruhiko, Hiraoka, Arudo, Sakaguchi, Taichi, Yoshitaka, Hidenori, Kitamura, Tetsuhisa, Komukai, Sho, Hirayama, Atsushi, Taniguchi, Kazuhiro, Sakata, Yasushi, and Sawa, Yoshiki

Published
2021
Full Text
View/download PDF

22. Real-world Instance-specific Image Goal Navigation for Service Robots: Bridging the Domain Gap with Contrastive Learning

Author

Sakaguchi, Taichi, Taniguchi, Akira, Hagiwara, Yoshinobu, Hafi, Lotfi El, Hasegawa, Shoichi, Taniguchi, Tadahiro, Sakaguchi, Taichi, Taniguchi, Akira, Hagiwara, Yoshinobu, Hafi, Lotfi El, Hasegawa, Shoichi, and Taniguchi, Tadahiro

Abstract

Improving instance-specific image goal navigation (InstanceImageNav), which locates the identical object in a real-world environment from a query image, is essential for robotic systems to assist users in finding desired objects. The challenge lies in the domain gap between low-quality images observed by the moving robot, characterized by motion blur and low-resolution, and high-quality query images provided by the user. Such domain gaps could significantly reduce the task success rate but have not been the focus of previous work. To address this, we propose a novel method called Few-shot Cross-quality Instance-aware Adaptation (CrossIA), which employs contrastive learning with an instance classifier to align features between massive low- and few high-quality images. This approach effectively reduces the domain gap by bringing the latent representations of cross-quality images closer on an instance basis. Additionally, the system integrates an object image collection with a pre-trained deblurring model to enhance the observed image quality. Our method fine-tunes the SimSiam model, pre-trained on ImageNet, using CrossIA. We evaluated our method's effectiveness through an InstanceImageNav task with 20 different types of instances, where the robot identifies the same instance in a real-world environment as a high-quality query image. Our experiments showed that our method improves the task success rate by up to three times compared to the baseline, a conventional approach based on SuperGlue. These findings highlight the potential of leveraging contrastive learning and image enhancement techniques to bridge the domain gap and improve object localization in robotic applications. The project website is https://emergentsystemlabstudent.github.io/DomainBridgingNav/., Comment: See website at https://emergentsystemlabstudent.github.io/DomainBridgingNav/. Submitted to IROS2024

Published
2024

27. Shape of the first mitotic spindles impacts nucleation in human embryos

Author

Ono, Yuki, primary, Shirasawa, Hiromitsu, additional, Kazumasa, Kazumasa, additional, Goto, Mayumi, additional, Ono, Takahiro, additional, Sakaguchi, Taichi, additional, Okabe, Motonari, additional, Hirakawa, Takeo, additional, Iwasawa, Takuya, additional, Fujishima, Akiko, additional, Onodera, Yohei, additional, Sugawara, Tae, additional, Makino, Kenichi, additional, Miura, Hiroshi, additional, Fukunaga, Noritaka, additional, Asada, Yoshimasa, additional, Kumazawa, Yukiyo, additional, and Terada, Yukihiro, additional

Published
2024
Full Text
View/download PDF

29. Mitral valve early systolic billowing induces following annular expansion and leaflet augmentation in Barlow's disease: sequential analysis using 3D echocardiography.

Author

Amano, Masashi, Izumi, Chisato, Toki, Misako, Yanagi, Yoshiki, Hayashida, Akihiro, Kawamoto, Takahiro, Hiraoka, Arudo, Fukushima, Satsuki, Sakaguchi, Taichi, Watanabe, Nozomi, and Yoshida, Kiyoshi

Subjects
MITRAL valve, MITRAL valve surgery, DATA analysis, MITRAL valve prolapse, DESCRIPTIVE statistics, MITRAL valve insufficiency, CARDIAC contraction, ANALYSIS of variance, STATISTICS, COMPARATIVE studies, ECHOCARDIOGRAPHY
Abstract

Aims Although systolic expansion of the annulus has been recognized in Barlow's disease, the mechanisms of the unique pathological movement of the annulus and its relation to the leaflet augmentation have not yet been clarified. We aimed to investigate the detailed mechanisms of the characteristic mitral apparatus dynamics in Barlow's disease by frame-by-frame sequential geometric analysis using real-time 3D transoesophageal echocardiography. Methods and results Fifty-three patients with Barlow's disease and severe mitral regurgitation without torn chordae, as well as 10 controls, were included. We evaluated geometric changes in the mitral complex using 3D transoesophageal echocardiography at five points during systole. To identify early systolic billowing of leaflets, the annulo-leaflet angle was measured. We also performed a more detailed analysis in four consecutive frames just before and after leaflet free-edge prolapse above the annulus plane. The median annulo-leaflet angle of both leaflets in early systole was >0° (above annulus plane) in patients with Barlow's disease, and billowing of the leaflet body was observed from early systole. The prolapse volume of both leaflets increased markedly from early to mid-systole [1.60 (0.85–2.80) to 4.00 (2.10–6.45) mL; analysis of variance (ANOVA), P < 0.001; post hoc , P < 0.05]. With frame-by-frame analysis, dynamic augmentation of the annulus and leaflets developed between frames just before and just after leaflet free-edge prolapse (ANOVA, P < 0.01; post hoc , P < 0.05). Conclusion In Barlow's disease, early systolic billowing of the mitral leaflet induces systolic annulus expansion followed by leaflet augmentation and leaflet free-edge prolapse. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

38. Mid-term outcomes of surgical aortic valve replacement using a mosaic porcine bioprosthesis with concomitant mitral valve repair

Author

Sugimori, Haruhiko, primary, Nakao, Tatsuya, additional, Okada, Yukikatsu, additional, Okita, Yutaka, additional, Yaku, Hitoshi, additional, Kobayashi, Junjiro, additional, Uesugi, Hideyuki, additional, Takanashi, Shuichiro, additional, Ito, Toshiaki, additional, Koyama, Tadaaki, additional, Sakaguchi, Taichi, additional, Yamamoto, Kouji, additional, Yoshikawa, Yasushi, additional, and Sawa, Yoshiki, additional

Published
2023
Full Text
View/download PDF

47. Development of Simple live-imaging method for viewing the first cleavage of mammalian embryos by using fluorescent chemical probes for DNA and cytoskeletons

Author

Okabe, Motonari, primary, Shirasawa, Hiromitsu, additional, Goto, Mayumi, additional, Iwasawa, Takuya, additional, Sakaguchi, Taichi, additional, Fujishima, Akiko, additional, Onodera, Yohei, additional, Makino, Kenichi, additional, Miura, Hiroshi, additional, Kumazawa, Yukiyo, additional, Takahashi, Kazumasa, additional, and Terada, Yukihiro, additional

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results