Kyutacar : Kyushu Institute of Technology Academic Repository
Not a member yet
9746 research outputs found
Sort by
Knowledge-Infused Retrieval Boosts Few-Shot Hand Gesture Recognition on HaGRID with Vision-Language Model
Get PDFHand gesture recognition remains challenging, primarily due to its reliance on large-scale annotated datasets and the limited adaptability of existing models when encountering novel gesture classes. In this work, we propose to apply Adaptive Vision-Language Model (Adaptive-VLM). This lightweight, training-free framework utilizes only one image per class to recognize gestures on the HaGRID benchmark. Built upon the CLIP backbone, our approach incorporates symbolic knowledge-infused prompts, multi-prompt contextualization, and semantic exemplar ranking to improve few-shot generalization. Adaptive-VLM achieves a macro F1-score of 65.75% on the HaGRID test set (540 images) without any parameter fine-tuning, using 18 example images. It significantly outperforms the Random-VLM baseline (59.95%) and a ResNet-18 model fine-tuned for 10 epochs (4.09%) under the same data constraints. These findings highlight the effectiveness of combining structured domain knowledge and guided exemplar selection to overcome data scarcity in low-resource gesture recognition. Adaptive-VLM offers a promising direction for building adaptive and efficient HGR systems, especially in real-world human-computer interaction scenarios requiring rapid deployment with minimal data.journal articl
Fuzzy Rule-Based Interpretation of Hand Gesture Intentions
Get PDFThis study investigates the interpretation of hand gestures in nonverbal communication, with particular attention paid to cases where gesture form does not reliably convey the intended meaning. Hand gestures are a key medium for expressing impressions, complementing or substituting verbal communication. For example, the “Thumbs Up” gesture is generally associated with approval, yet its interpretation can vary across contexts and individuals. Using participant-generated descriptive words, sentiment analysis with the VADER method, and fuzzy membership modeling, this research examines the variability and ambiguity in gesture–intention mappings. Our results show that Negative gestures, such as “Thumbs Down,” consistently align with Negative sentiment, while Positive and Neutral gestures, including “Thumbs Sideways” and “So-so,” exhibit greater interpretive flexibility, often spanning adjacent sentiment categories. These findings demonstrate that rigid, category-based classification systems risk oversimplifying nonverbal communication, particularly for gestures with higher interpretive uncertainty. The proposed fuzzy logic-based framework offers a more context-sensitive and human-aligned approach to modeling gesture intention, with implications for affective computing, behavioral analysis, and human–computer interaction.journal articl
Creating a Future Space Where Humans and Robots Live Together
Get PDF本稿では2023年4月15日から5月28日にかけて募集したクラウドファンディング（CF）プロジェクト「人とロボットが一緒に暮らす未来の空間を創る」[1] の取り組みについて紹介する．journal articl
An Automatic Security Monitoring System with Single Optical Sensor for Home Appliances
Get PDFAging will become a significant societal challenge on a global scale, household appliances monitoring plays a significant role in home safety, especially for the elderly. This study introduces a unified monitoring system designed to oversee nine critical household events: lighting, TV, stove, air conditioner, water supply, door lock, window lock, heater, and bath. The system utilizes a single photodiode optical sensor and infrared LEDs to monitor these events, eliminating the need for multiple sensors and complex installations. We innovatively proposed the integration of a CPU-based noise filter, which significantly improved the stability and accuracy of data acquisition. Additionally, the temperature-sensitive properties of infrared LEDs were harnessed for precise state identification of each monitored event. We propose future enhancements to improve the system's sensitivity to minor temperature variations by addressing voltage fluctuations caused by the delayed response of LED elements to temperature changes. By refining the threshold settings, the system could provide more detailed assessments, making it applicable to a broader range of household monitoring scenarios. The experimental results confirmed that this approach effectively monitors all nine events with a single device, demonstrating the system's efficiency and practicality. This research lays the groundwork for more comprehensive and cost-effective smart home systems that simplify the management of various household events while maintaining high accuracy and reliability.journal articl
A study of the finite element method employing three-node elements in a one-dimensional domain
Get PDFThree-node isoparametric elements are widely used for solving various one-dimensional problems with finite element discretization. However, it is not well known that the value of the Jacobian at either end of an element can be zero or negative when the interior node is placed at particular positions inside the element. In this study, we propose a simple remedy to fix this problem, and the proposed method is applied to the discretization of the one-dimensional advection-diffusion equation. With the use of this proposed scheme, accurate numerical results are obtained on two types of unequally spaced meshes, whereas the numerical solution obtained using isoparametric elements diverges from the analytical solution on one of the meshes, whose nodes are more concentrated in the vicinity of the edges of the domain. This proves the robustness of the proposed method on any unequally divided mesh.journal articl
Range of effectiveness of hydraulic diameter model as an analytical solution for rectangular microchannels
Get PDFSpontaneous capillary flow in rectangular microfluidic channels is employed in microfluidic devices for various applications. The exact solution for flow in a rectangular cross-sectional channel has a complex point that contains an infinite sum term. The flow depends on the depth-width ratio of the rectangular channel's cross-section, ε. In previous studies, several approximations from exact solutions were useful for ε values smaller than 1 or 2. In this study, we propose a conversion equation J(ε) that turns the hydraulic-diameter (HD) model into a rectangular-channel (RC) model. Experiments: Rectangular type flow experiments with ethanol solution were conducted to confirm the difference between the RC and HD models when the analysis was performed based on each model. Findings: J(ε) depends on ε. At ε=0.441, 2.27, the two models coincide. For 0.247 < ε < 4.04, J(ε) deviates by ±12.5.journal articl
Dynamic Hand Gesture Recognition by Hand Landmark Classification Using Long Short-Term Memory
Get PDFHand gestures are a valuable modality for human-computer interaction, conveying information that can be used as input. Dynamic hand gestures, prevalent in real-world scenarios, necessitate considering temporal factors such as gesture initiation, termination, and frame sequence. A Long Short-Term Memory (LSTM) based recognition model was proposed to address this challenge. Data availability for dynamic hand gesture research is a significant hurdle. The dataset introduced by Fronteddu et al. provides 27 classes of dynamic hand gestures, serving as a suitable training resource. MediaPipe Hands, a computer vision framework, was leveraged to extract keypoints from each frame, capturing spatial features fed into the LSTM model. Experiments were conducted to determine the optimal dropout rate for the LSTM model. Results indicated that a dropout rate of 70% yielded the highest accuracy, achieving up to 98.53% validation accuracy and 99.71% test accuracy. These findings demonstrate the effectiveness of the proposed LSTM-based recognition model for dynamic hand gestures. Future research could explore integrating other deep learning techniques, such as attention mechanisms, to enhance the accuracy and robustness of dynamic hand gesture recognition systems. Additionally, investigating the application of the proposed model in real-world scenarios, such as virtual and augmented reality, would be valuable in assessing its practical utility.journal articl
Multiple anisotropy of cellulose nanocrystals self-organized into disc-shaped particles
No full textWe fabricated a disc-shaped particulate composite based on cellulose nanocrystals (CNCs) and polyethylene glycol diacrylate (PEGDA) via a self-organized precipitation method. The disc-shaped particles had a structure with CNCs segregated in the outer region. The disc-shaped particulate composite also shows a higher refractive index in the direction perpendicular to the disc surface. This means that this particle system has “double” anisotropy: spatially and optically. Moreover, the disc orientation can be manipulated by the application of AC.journal articl