HTHaruo Takemura

Haruo Takemura

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Haruo Takemura has a total of 76 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • TransPortal: A Portable Portal-Based Locomotion Technique

    by Daichi Hirobe, Shizuka Shirai, Jason Orlosky, Masato Kobayashi, Yuki Uranishi, Haruo Takemura

    In VR video games, users can explore virtual worlds by physically walking in the real world or by using alternative locomotion techniques such as teleportation or steering. However, these techniques often pose challenges, including disorientation or sickness. This study presents TransPortal, a novel portal-based locomotion technique for VR video games. The proposed technique allows users to manipulate a marker object to specify a destination and to move there through a portal that can be freely activated or deactivated. To evaluate the proposed method, we conducted a within-subjects experiment with 30 participants, comparing it with traditional teleportation and steering. The results show that TransPortal induces lower VR sickness in a complex scenario without reducing immersion or entertainment quality.

  • CaliView: Continuous Viewpoint Calibration Using Dynamic Rotation Gain Control

    by Donghae Lim, Shizuka Shirai, Masato Kobayashi, Yuki Uranishi, Haruo Takemura

    Head tracking allows users of Virtual Reality (VR) to freely rotate their heads 360 degrees while exploring virtual environments. When using VR in a limited space, the ability to physically rotate one’s head is limited to a specific range. To address this issue, previous studies have proposed methods to employ distinct rotation factors for real and imaginary rotations. However, its primary usage lies in redirected walking; thus, it is unsuitable for seated VR. In this article, we propose CaliView, which consistently adjusts the user’s perspective to always face an optimal direction in VR, all while ensuring a comfortable posture. CaliView continuously controls the rotation gain to ensure that the disparity between the present body orientation and the optimal orientation is reduced to zero, encouraging the user to assume a forward-facing position with the target orientation. To assess the suggested approach, CaliView, we experimented to compare three conditions: CaliView, snap turning only (SnapTurn), and hybrid of CaliView and snap turning (Hybrid). The research findings suggest that CaliView functions as a useful reorientation technique, enabling implicit reorientation without sacrificing the user experience. Additionally, this study showcases its compatibility with various other techniques, such as the traditional snap-turn, thus emphasizing its versatility.

  • Open Access

    User-Centric Locomotion Techniques for Virtual Reality Games: A Survey of User Needs and Issues

    by Daichi Hirobe, Shizuka Shirai, Jason Orlosky, Mehrasa Alizadeh, Masato Kobayashi, Yuki Uranishi, Photchara Ratsamee, Haruo Takemura

    Virtual reality (VR) video games that are played on a VR headset are becoming increasingly common in households, and though many games require players to navigate vast virtual spaces, most homes cannot provide a large enough physical space to encompass the entire virtual space. Thus, VR video games that require locomotion often provide users with alternative locomotion techniques. While teleportation or steering is typically used as a standard, new techniques can overcome remaining problems such as motion sickness. However, a holistic perspective of user needs and issues regarding these techniques in practical situations has not been studied on a broad basis. To address this gap in the literature and contribute to future VR video game development and research, we conducted 16 semi-structured interviews and surveyed 88 participants to help explore issues regarding existing locomotion techniques. Our results revealed preferences related to teleportation versus steering and the postures that users adopt while playing VR video games, along with user needs for locomotion techniques in each posture.


  • Open Access

    Cloud Computing Challenges and Needs in Higher Education Institutions in Post-COVID-19 Times: A Case of a Japanese Survey

    by David Soto, Shizuka Shirai, Mayumi Ueda, Manabu Higashida, Yuki Uranishi, Haruo Takemura

    Cloud computing has become essential for Higher Education Institutions (HEIs) due to its high availability, scalability, and support for virtual learning environments. The COVID-19 pandemic further accelerated this trend, prompting HEIs to transition their learning environments to the cloud. Consequently, this rapid shift has introduced post-adoption challenges, including security and privacy concerns, compliance, cost management, technical expertise, and the necessity of reliable Internet connectivity. Previous research has primarily focused on frameworks for cloud adoption in HEIs. However, there is a gap in studies addressing post-adoption challenges. Particularly in Japanese higher education, despite high adoption rates, HEIs are still facing institutional and technical difficulties in effectively leveraging cloud benefits. To address this gap, this paper presents an exploratory study conducted in 97 Japanese HEIs through a questionnaire based on the Technology-Organization-Environment (TOE) framework to examine the cloud post-adoption benefits and challenges. The survey assessed nine factors and various institutional attributes. Findings confirmed a high adoption level in Japanese HEIs (99%), with Hybrid Cloud (56%) and Software as a Service (SaaS) (96%) being prevalent. COVID-19 significantly influenced cloud adoption, with 92% of HEIs acknowledging increased awareness and use of cloud technologies, and 70% recognizing that cloud-based apps facilitated more efficient learning. Nonetheless, data security remains a concern, with 55% of HEIs worried about data leakage, unauthorized access, and storing critical information in the cloud. Future initiatives should address reported barriers such as budget management within the Pay-as-you-go model, lack of expertise, and the development of frameworks for continuous cloud operations in HEIs.


  • Open Access

    Identifying Disaster Regions in Images Through Attention Shifting With a Retarget Network

    by Narongthat Thanyawet, Photchara Ratsamee, Yuki Uranishi, Masato Kobayashi, Haruo Takemura

    Disasters disrupt lives and necessitate quick location of affected areas for rescue efforts. The application of computer vision has enhanced disaster detection, such as landslides and floods; however traditional computer vision methods often overlook smaller, critical details in favor of prominent objects. This research introduces the Retarget Network (RetNet), a novel framework aimed at improving image captioning techniques to identify and prioritize these less evident, yet crucial, objects in disaster scenarios, enhancing scene recognition and aiding in more effective disaster response. By masking images, we direct the model’s focus towards additional significant areas within the image. RetNet employs anchor boxes to refine the targeting of specific areas and, optimize their center positions, heights, and widths. Additionally, RetNet determines which anchors to mask prior to captioning, facilitating the identification of challenging objects such as boulders, soil, and water, which are indicative of natural disasters. We validated RetNet across multiple disaster scenarios—landslides, floods, and wildfires—using images taken from various perspectives, including side-view, aerial and shipborne views. Our findings reveal an accuracy of 91.60% in landslide detection from side-view image captions and 87.50% for detections from shipborne views. These results underscore RetNet’s effectiveness in enhancing the identification of disaster-affefted regions.


  • Open Access

    Generation of Inverted Locomotion Gait for Multi-Legged Robots Using a Spherical Magnetic Joint and Adjustable Sleeve

    by Harn Sison, Photchara Ratsamee, Manabu Higashida, Yuki Uranishi, Haruo Takemura

    Title: Generation of Inverted Locomotion Gait for Multi-Legged Robots Using a Spherical Magnetic Joint and Adjustable Sleeve | Keywords: inverted locomotion robot, multi-legged robot, inverted locomotion gait pattern, permanent magnet, adjustable sleeve | Author: Harn Sison, Photchara Ratsamee, Manabu Higashida, Yuki Uranishi, and Haruo Takemura


  • Panoptic-Level Image-to-Image Translation for Object Recognition and Visual Odometry Enhancement

    by Liyun Zhang, Photchara Ratsamee, Zhaojie Luo, Yuki Uranishi, Manabu Higashida, Haruo Takemura

    Image-to-image translation methods have progressed from only considering the image-level information to integrating the global- and instance-level information. However, only the foreground instances are refined, and the background semantics are taken as an entire feature, which causes a substantial loss of the semantic information in the translation. Additionally, the insufficient quality of the translated semantic regions also leads to an unsatisfactory performance of the object recognition or visual odometry tasks in which the translated images/videos are further used. In this paper, we propose a novel generative adversarial network for panoptic-level image-to-image translation (PanopticGAN). The proposed method has three advantages: (1) the extracted panoptic perception (i.e., the foreground instances and background semantic regions) as content codes are aligned with the sampled panoptic style codes, which considers the panoptic-level information to avoid the semantic information loss, and the latent space of each object has a rich fusion of content and style codes to generate the higher-fidelity results; (2) a feature masking module is proposed to extract the representations within each object contour by masks for sharpening the object boundaries; (3) the improved fidelity of the translated semantic regions further contributes to enhancing the performance of the object recognition or visual odometry tasks that the translated images/videos are used in. In this paper, we also annotate a compact panoptic segmentation dataset for the thermal-to-color translation task. Extensive experiments are conducted to demonstrate the effectiveness of our PanopticGAN over the latest methods.


  • Open Access

    Multi-modal humor segment prediction in video

    by Zekun Yang, Yuta Nakashima, Haruo Takemura

    Humor can be induced by various signals in the visual, linguistic, and vocal modalities emitted by humans. Finding humor in videos is an interesting but challenging task for an intelligent system. Previous methods predict humor in the sentence level given some text (e.g., speech transcript), sometimes together with other modalities, such as videos and speech. Such methods ignore humor caused by the visual modality in their design, since their prediction is made for a sentence. In this work, we first give new annotations to humor based on a sitcom by setting up temporal segments of ground truth humor derived from the laughter track. Then, we propose a method to find these temporal segments of humor. We adopt an approach based on sliding window, where the visual modality is described by pose and facial features along with the linguistic modality given as subtitles in each sliding window. We use long short-term memory networks to encode the temporal dependency in poses and facial features and pre-trained BERT to handle subtitles. Experimental results show that our method improves the performance of humor prediction.


  • Open Access

    SmartVP: Viewpoint Optimization Based on Individual Preference for Watching 3D Boxing Punch Videos

    by Tao Tao, Photchara Ratsamee, Chang Liu, Yuki Uranishi, Haruo Takemura

    When broadcasting the 3D boxing punch videos, spectators may have different preferences for spectating. Some people prefer to see the hitting point, while others want to see the punching form. It is not easy to provide a uniformly optimal viewpoint. In this paper, we introduce a novel method to select an optimal viewpoint for watching punching moments in VR that can handle different spectator preferences. We customize a visibility model that utilizes 8 bounding boxes to account for the visibility of upper body parts. We use a neural network classification model to reproduce the optimal viewpoint selection based on the features of body parts visibility, punch side, and punch offset. We collected spectators’ preferred viewpoints from 24 participants under 3 controlled preference conditions: seeing the punch Arm Form clearly (AF), seeing the Facial Expression clearly (FE), and no additional restriction (None). We conducted a user study to evaluate the empirical performance of our system. The training and user study results show that our method can reproduce the optimal viewpoint with spectator preference.

  • Open Access

    Evaluation of User Interfaces for Three-Dimensional Locomotion in Virtual Reality

    by Donghae Lim, Shizuka Shirai, Jason Orlosky, Photchara Ratsamee, Yuki Uranishi, Haruo Takemura

    Locomotion is an essential factor for interaction in virtual environments. Virtual reality allows users to move freely from the constraints of gravity or the environment, which is impossible in the real world. These experiences, in which users can move at all 6 degrees of freedom, can enable new navigation paradigms that exceed 2D ground-based movement in terms of enjoyment and interactivity. However, most existing VR locomotion interfaces have limitations because they are developed for ground-based locomotion, constraining the degrees of freedom (DoF) during movement as in the real world. This exploratory study was designed to seek the features required for three-dimensional (3D) locomotion by evaluating three types of interfaces: Slider, Teleport, and Point-Tug, which were implemented based on existing ground-based locomotion techniques. We then conducted a user study with 3D navigation tasks, using both objective and subjective measures to evaluate efficiency, overall usability, motion sickness, perceived workload, and presence. The results suggest that Slider has an advantage compared to the others in terms of usability and perceived workload since it can freely and intuitively designate the direction of movement.