The importance of a high-quality dataset availability in 3D human action analysis research cannot be overstated. This paper introduces DGU-HAO (Human Action analysis dataset with daily life Objects). This novel 3D human action multi-modality dataset encompasses four distinct data modalities accompanied by annotation data, including motion capture data, RGB video data, image data, and 3D object modeling data. It features 63 action classes involving interactions with 60 common furniture and electronic devices. Each action class comprises approximately 1, 000 motion capture data representing 3D skeleton data, along with corresponding RGB video and 3D object modeling data, resulting in 67, 505 motion capture data samples. It offers comprehensive 3D structural information of the human, RGB images and videos, and point cloud data for 60 objects, collected through the participation of 126 subjects to ensure inclusivity and account for diverse human body types. To validate our dataset, we leveraged MMNet, a 3D human action recognition model, achieving Top-1 accuracy of 91.51% and 92.29% using the skeleton joint and bone methods, respectively. Beyond human action recognition, our versatile dataset is valuable for various research endeavors in 3D human action analysis.

There is a growing need for systems that can be used to effectively detect and classify intrusions in extensive network data exchanges. To this end, we propose Hi-MLIC, a hierarchical multilayer lightweight intrusion classification model that has been designed to address various intrusion types. This study highlights the challenges involved in classifying intrusions due to data imbalance across different types of intrusion data along with the complex nature of consolidating multiple benchmark datasets into cohesive datasets for real-time detection. To address these issues, we consolidated packet capture data from two widely used benchmark datasets, CIC-IDS2017 and UNSW-NB15, into two newer and more comprehensive datasets, CM-CIC-IDS2017 and CM-UNSW-NB15, respectively. This consolidation enables the identification and classification of a broader range of intrusion types. Our hierarchical approach achieves improved classification accuracy by effectively addressing the class imbalance that is inherent in non-hierarchical models. Layer-1 separates network traffic into benign and malicious categories. Layer-2 further classifies malicious traffic into four groups, while Layer-3 identifies 23 specific intrusion types. We reduced the model complexity and processing time by performing misclassification analysis and eliminating unnecessary features. Our model ultimately achieved a recall metric of up to 98.8%, thus demonstrating its effectiveness and efficiency in intrusion detection and classification. Altogether, the proposed Hi-MLIC represents a significant advancement in addressing the challenges of real-time network intrusion detection.

The importance of a high-quality dataset availability in 3D human action analysis research cannot be overstated. This paper introduces DGU-HAO (Human Action analysis dataset with daily life Objects). This novel 3D human action multi-modality dataset encompasses four distinct data modalities accompanied by annotation data, including motion capture data, RGB video data, image data, and 3D object modeling data. It features 63 action classes involving interactions with 60 common furniture and electronic devices. Each action class comprises approximately 1, 000 motion capture data representing 3D skeleton data, along with corresponding RGB video and 3D object modeling data, resulting in 67, 505 motion capture data samples. It offers comprehensive 3D structural information of the human, RGB images and videos, and point cloud data for 60 objects, collected through the participation of 126 subjects to ensure inclusivity and account for diverse human body types. To validate our dataset, we leveraged MMNet, a 3D human action recognition model, achieving Top-1 accuracy of 91.51% and 92.29% using the skeleton joint and bone methods, respectively. Beyond human action recognition, our versatile dataset is valuable for various research endeavors in 3D human action analysis.

There is a growing need for systems that can be used to effectively detect and classify intrusions in extensive network data exchanges. To this end, we propose Hi-MLIC, a hierarchical multilayer lightweight intrusion classification model that has been designed to address various intrusion types. This study highlights the challenges involved in classifying intrusions due to data imbalance across different types of intrusion data along with the complex nature of consolidating multiple benchmark datasets into cohesive datasets for real-time detection. To address these issues, we consolidated packet capture data from two widely used benchmark datasets, CIC-IDS2017 and UNSW-NB15, into two newer and more comprehensive datasets, CM-CIC-IDS2017 and CM-UNSW-NB15, respectively. This consolidation enables the identification and classification of a broader range of intrusion types. Our hierarchical approach achieves improved classification accuracy by effectively addressing the class imbalance that is inherent in non-hierarchical models. Layer-1 separates network traffic into benign and malicious categories. Layer-2 further classifies malicious traffic into four groups, while Layer-3 identifies 23 specific intrusion types. We reduced the model complexity and processing time by performing misclassification analysis and eliminating unnecessary features. Our model ultimately achieved a recall metric of up to 98.8%, thus demonstrating its effectiveness and efficiency in intrusion detection and classification. Altogether, the proposed Hi-MLIC represents a significant advancement in addressing the challenges of real-time network intrusion detection.

Eyebrows are the most important facial feature in facial recognition with its shape rated to be more helpful than color or density for facial reconstruction or approximation. However, little extant research has estimated the position and morphological territory of the eyebrow from the orbit. Three-dimensional craniofacial models, produced from CT scans of 180 Koreans autopsied at the National Forensic Service Seoul Institute, were used to conduct metric analyses of subjects (125 males and 55 females) between 19 and 49 (mean 35.1) years. We employed 18 craniofacial landmarks to examine the morphometry of the eyebrow and orbit with 35 pairs of distances per subject measured between landmark and reference planes. Additionally, we used linear regression analyses to predict eyebrow shape from the orbit for every possible combination of variables. The morphology of the orbit has more influence on the position of the superior margin of the eyebrow. In addition, the middle part of the eyebrow was more predictable. The highest point of the eyebrow in female was located more medially than the male. Based on our findings, the equations for estimating the position of the eyebrow from the shape of the orbit is useful information for face reconstruction or approximation.

This paper proposes a method for selecting commutation intervals and a novel implementation approach for motor drives by using a thyristor-based load commutated inverter (LCI) during forced commutation operation. The equivalent circuit of the LCI system is modeled during forced commutation, and the current and voltage characteristics are formulated into ignition equations to develop a stable method for selecting commutation intervals. Additionally, an algorithm is proposed to implement and maintain the selected commutation intervals throughout operation. The proposed method for selecting commutation intervals and the corresponding implementation algorithm are verified through MATLAB / SIMULINK and experiments using a scaled model.

[목적] 본 연구는 수어가 고유한 문법과 양식을 지닌 농인의 언어이며, 수어통역에서도 음성 언어와 동일한 방법과 메커니즘(mechanism)이 적용할 수 있다는 것을 토대로 수어 동시통역 전략의 사례를 분석하였다. 그간 수어통역에 대한 품질 평가관점에서 연구가 이루어져, 통역 과정에 대한 체계적인 방법론을 제시하지 못해 왔다. 그리하여 수어통역 현장에서의 실천적 대안과 방법을 제시하고자 수어의 시각언어 특성을 반영한 ‘수어 동시통역 전략의 사례’를 분석하여 연구하고자 한다. [방법] 연구방법은 수어통역사 5명을 대상으로 한국어(출발어, L1)에서 수어(도착어, L2)로 동시통역을 촬영하였으며, 이후 통역 결과물은 농인 3명에 의해 의미이해 평가를 받았다. 전략의 사례를 분석하는 과정에서 여러 통역 전략들이 상호 관계 속에서 네트워크가 이루어지는 등 구분의 어려움이 있으나, 전략의 개념에 더욱 일치한 부분을 구분하여 사례 분석하였다. [결과] 사례분석 결과를 토대로, 시각언어인 수어의 특성을 반영한 전략 유형으로 다섯 가지를 제시한다. 첫째, 메시지 이해를 위한 인지 전략으로는 EVS(Ear-Voice Span, 청성시차), 휴지(Pause), 의미덩이 짓기(Chunking, 대기:Waiting), 예측(Anticipation)이 있다. 둘째는 의미 효과를 위한 생산 전략으로 통사구조 변환(Syntactic Transformation), 단어 및 구절 대체(Substitution), 구체화(Particularization), 추가(Addition), 부가설명(Additional Explanation)이 있다. 셋째는 표현의 경제성 전략으로 생략(Omission), 요약(Summarizing), 압축(Compression), 일반화(Generalization)가 있다. 넷째는 비상 전략으로 직역(Transcoding)과 누락(Omission)이 있다. 다섯째는 메타(Meta) 전략으로의 모니터링(Monitoring)은 수정(Repair)이 있다. [결론] 본 연구의 사례분석 및 전략유형을 통해 수어 동시통역의 실질적인 방법론을 제시함으로써, 수어 통역 방법의 질적인 향상에 기여하고자 한다.

Constructing diverse and complex multi-modal datasets is crucial for advancing human action analysis research, providing ground truth annotations for training deep learning networks, and enabling the development of robust models across real-world scenarios. Generating natural and contextually appropriate nonverbal gestures is essential for enhancing immersive and effective human–computer interactions in various applications. These applications include video games, embodied virtual assistants, and conversations within a metaverse. However, existing speech-related human datasets are focused on style transfer, so they have limitations that make them unsuitable for 3D human action analysis studies, such as human action recognition and generation. Therefore, we introduce a novel multi-modal dataset, DGU-HAU, a dataset for 3D human action on utterances that commonly occurs during daily life. We validate the dataset using a human action generation model, Action2Motion (A2M), a state-of-the-art 3D human action generation model.

Humans can recognize or infer unseen classes of objects using descriptions explaining the characteristics (semantic information) of the classes. However, conventional deep learning models trained in a supervised manner cannot classify classes that were unseen during training. Hence, many studies have been conducted into generalized zero-shot learning (GZSL), which aims to produce system which can recognize both seen and unseen classes, by transferring learned knowledge from seen to unseen classes. Since seen and unseen classes share a common semantic space, extracting appropriate semantic information from images is essential for GZSL. In addition to semantic-related information (attributes), images also contain semantic-unrelated information (contents), which can degrade the classification performance of the model. Therefore, we propose a content-attribute disentanglement architecture which separates the content and attribute information of images. The proposed method is comprised of three major components: 1) a feature generation module for synthesizing unseen visual features; 2) a content-attribute disentanglement module for discriminating content and attribute codes from images; and 3) an attribute comparator module for measuring the compatibility between the attribute codes and the class prototypes which act as the ground truth.With extensive experiments, we show that our method achieves state-of-the-art and competitive results on four benchmark datasets in GZSL. Our method also outperforms the existing zero-shot learning methods in all of the datasets. Moreover, our method has the best accuracy as well in a zero-shot retrieval task.

본 논문에서는 OpenCV와 MediaPipe를 활용하여 실시간 영상의 운동 자세(스쿼트, 팔굽혀펴기)를 추정해 운동 자세교정 내용과 운동횟수를 안내하는 프로그램을 제안한다. 구체적으로 프로그램은 실시간으로 웹캠을 이용하여 획득한 영상에서 사용자 신체의 landmark를 추정하고, 올바른 자세 여부 판단에 필요한 몸의 각도, 수치를 계산하여 사용자가 잘못된 자세를 취했을 때 교정에 필요한 내용을 안내할 수 있다. 또한, 사용자가 올바른 자세를 취했을 때 운동횟수를 증가시키고 안내한다. 자세교정을 위한 안내 문구가 화면에 표시되고 운동횟수가 화면과 음성으로 안내된다. 이 프로그램은 집에서 운동 트레이너의 도움 없이 자신의 자세를 확인하면서 올바른 자세로 운동할 수 있도록 돕는다.