Sanghoon Lee

The basketball jump shot is a fundamental skill in competitive play, and defensive pressure has been shown to influence movement execution and post-landing control. This study investigated post-landing stability during basketball jump shots performed under virtual reality (VR) defense, real defense, and no-defense conditions. Nineteen basketball club members performed jump-shot landings while ground reaction forces were recorded using a force plate. Time to stabilization (TTS), center of pressure (COP) path length, and jump height were analyzed. The presence of defensive cues was associated with shorter TTS than in the no-defense condition (p < .05), whereas no significant difference was observed between the VR and real defense conditions. COP path length did not differ significantly across conditions. However, jump height was significantly lower in the VR defense condition than in both the real defense and no-defense conditions (p < .05). These findings suggest that VR-based defensive stimuli can alter movement execution and landing responses but do not fully replicate the biomechanical demands of real defensive interactions. Rather, VR represents a distinct sensorimotor context that shapes motor performance. Although VR provides a structured and sport-relevant environment for assessing landing stability, findings obtained in virtual settings should not be interpreted as direct equivalents of real-world performance. Further research incorporating broader biomechanical and perceptual measures is needed to clarify the relationship between VR-based assessments and real-world basketball performance.

Light field (LF) imaging is inherently constrained by the trade-off between spatial resolution and angular sampling density. To overcome this obstacle, spatial-angular super-resolution (SR) methods have been developed to achieve concurrent enhancement in both dimensions. Traditional spatial-angular SR methods treat spatial and angular SR as separate tasks, resulting in parameter redundancy and error accumulation. While recent end-to-end approaches attempt joint processing, their uniform treatment of these distinct problems overlooks critical domain-specific requirements. To address these challenges, we propose a domain-specialized framework that deploys stage-tailored strategies to satisfy domain-specific demands. Specifically, in the angular SR stage, we introduce a cross-view consistency modulation module that enhances inter-view coherence through long-range dependency modeling of angular features. In the spatial SR stage, we propose a detail-aware state space model to reconstruct fine-grained detail. Finally, we develop a cross-domain integration module that explores spatial-angular correlations by fusing multi-representational features from both domains to foster synergistic optimization. Experimental results on public LF datasets demonstrate substantial improvements over state-of-the-art methods in both qualitative and quantitative comparisons, with approximately 50% fewer model parameters compared to competing methods.

In this study, a conjugated heat transfer (CHT) analysis was performed to investigate the thermal environment characteristics of heat exchangers and the structure in a scramjet combustor.The combustor model was a kerosene-fueled tandem cavity scramjet combustor with a heat exchanger installed on both the upper and lower panels.A pseudo-CHT (p-CHT) analysis algorithm was proposed, decoupling the combustion simulation for the flow field and the heat transfer analysis for the solid structure to account for the ground combustion test time, which is on the order of tens of seconds.Combustion simulation was conducted based on the Improved Delayed Detached Eddy Simulation (IDDES) turbulence model, applying a global chemical reaction mechanism for surrogate kerosene.Wall temperature data obtained from the combustion simulation were utilized as boundary conditions for heat flux in the p-CHT analysis, and data for approximately 60 seconds were acquired.In the p-CHT analysis, the heat flux was concentrated on the rear wall region of the first cavity on the upper panel, revealing a maximum temperature distribution of approximately 1,040 K. On the lower panel, a comparatively lower heat flux was concentrated, yielding a maximum temperature peak of approximately 850 K.The temperature gradient between the upper and lower panels indicates the potential for structural stability concerns in the heat exchanger and jacket.And the location of the maximum temperature derived from the p-CHT analysis did not coincide with that from the combustion simulation.These findings suggest that the structural thermal environment is generated through the coupling of flame structure and combustor geometry conditions.Through these series of results, it was found that the proposed p-CHT analysis algorithm is capable of conducting conjugated heat transfer analysis for durations of tens of seconds with notably low computational cost.

Modeling relightable and animatable human avatars from monocular video is a long-standing and challenging task. Recently, Neural Radiance Field (NeRF) and 3D Gaussian Splatting (3DGS) methods have been employed to reconstruct the avatars. However, they often produce unsatisfactory photo-realistic results because of insufficient geometrical details related to body motion, such as clothing wrinkles. In this paper, we propose a 3DGS-based human avatar modeling framework, termed as Relightable and Dynamic Gaussian Avatar (RnD-Avatar), that presents accurate pose-variant deformation for high-fidelity geometrical details. To achieve this, we introduce dynamic skinning weights that define the human avatar's articulation based on pose while also learning additional deformations induced by body motion. We also introduce a novel regularization to capture fine geometric details under sparse visual cues. Furthermore, we present a new multi-view dataset with varied lighting conditions to evaluate relight. Our framework enables realistic rendering of novel poses and views while supporting photo-realistic lighting effects under arbitrary lighting conditions. Our method achieves state-of-the-art performance in novel view synthesis, novel pose rendering, and relighting.

In this demonstration, we present Permission to Dance, an end-to-end dance enhancement system designed to capture, enhance, and analyze user's dance performance. Our system consists of a dance capture module, a dance enhancement module, and a dance feedback module. Using the system, users can acquire their dance data in an enhanced version, followed by textual feedback on how to achieve better dance performance. The demonstration video is available at https://youtu.be/lFw7Xic48KU

Non-rigid registration is essential for reconstructing dynamic and incomplete 3D human meshes, yet traditional methods often fail to achieve robust alignment in the sequence of high-motion deformations and missing geometry. We propose a domain crossover non-rigid registration (DCNRR) framework that addresses these challenges by effectively transferring informative features from 2D image space into the 3D mesh domain of three key stages: multi-view projection, hierarchical non-rigid registration, and topology-consistent completion. In the first stage, multi-view projections are used to extract 2D joint locations and deep features, which guide deformation in the 3D space. In the second stage, hierarchical joint priors and deep features collaboratively guide mesh alignment, enabling more accurate deformation in distal regions and complex poses. In the final stage, we apply a diffusion-based completion process in UV coordinates to reconstruct incomplete surface normals and refine missing mesh areas with topological consistency. Our approach achieves highly detailed and perceptually accurate mesh deformation. To validate our approach, we evaluate performance on a newly constructed dynamic human motion (DHM) dataset, as well as public datasets. Our method demonstrates state-of-the-art results in both geometric accuracy and stability, showing particular robustness in dynamic and incomplete mesh sequences.

Tire wear particles (TWP) are increasingly recognized as a major contributor to urban road dust and an emerging environmental risk; however, their year-round dynamics and removal mechanisms remain poorly constrained. We developed a probabilistic simulation framework to evaluate TWP concentrations across urban arterial, industrial, and residential roads in Seoul. The model incorporated stochastic rainfall wash-off and probabilistic cleaning events, and the results were validated against measured TWP concentrations. Modeled uncertainty ranges overlapped with field observations, with residential roads showing the closest agreement, while urban and industrial roads were underestimated but remained within observed ranges. Results revealed clear road-type and seasonal dependencies: the lowest concentrations were found on urban roads due to frequent cleaning, the highest concentrations were found on industrial roads with wide variability, and residential roads exhibited strong sensitivity to rainfall, showing sharp declines during the monsoon. Scenario analyses further demonstrated the complementary roles of rainfall and cleaning: when cleaning was eliminated, TWP concentrations markedly increased; enhancing the cleaning efficiency reduced accumulation year-round; the no-rainfall case underscored the irreplaceable role of precipitation in producing the lowest TWP concentrations during summer; and a climate-change scenario with 1.5-fold rainfall indicated divergent road-type responses.

Robot Operating System 2 (ROS 2) relies on the Data Distribution Service (DDS), which offers more than 20 Quality of Service (QoS) policies governing availability, reliability, and resource usage. Yet ROS 2 users lack clear guidance on safe policy combinations and validation processes prior to deployment, which often leads to trial-and-error tuning and unexpected runtime failures. To address these challenges, we analyze DDS Publisher-Subscriber communication over a life cycle divided into Discovery, Data Exchange, and Disassociation, and provide a user oriented tutorial explaining how 16 QoS policies operate in each phase. Building on this analysis, we derive a QoS dependency chain that formalizes inter-policy relationships and classifies 41 dependency violation rules, capturing constraints that commonly cause communication failures in practice. Finally, we introduce QoS Guard, a ROS 2 package that statically validates DDS XML profiles offline, flags conflicts, and enables safe, predeployment tuning without establishing a live ROS 2 session. Together, these contributions give ROS 2 users both conceptual insight and a concrete tool that enables early detection of misconfigurations, improving the reliability and resource efficiency of ROS 2 based robotic systems.

This study conducted an experimental investigation to evaluate the effects of injector geometries on the combustion characteristics of a kerosene-fueled tandem cavity direct-connect supersonic combustor with regenerative cooling channels.A high-enthalpy flow at Mach number of 2.0 was supplied using both electric and vitiated air heaters.Four injector configurations, with varying exit areas and numbers, all maintaining the same mass flow rate, were utilized.The flame stabilization and combustion performance of each injector configurations were classified through quantitative analysis using wall static pressure distribution, specific fuel consumption rate, and fuel heating, and qualitative analysis through flame visualization images.Additionally, it is suggested that an increase in the heat exchanger surface area and modifications to the cooling channels are necessary to anchor a stable flame under higher fuel flow conditions. . 2.0 .

Microvascular decompression (MVD) is a neurosurgical procedure that relieves nerve compression by repositioning or separating offending blood vessels, effectively reducing pain or spasms. Accurate localization of the compression site is crucial for optimal surgical outcomes, as it enables precise identification and decompression of the offending vessel. While horizontal anatomical relationships are easily identified in the surgical view, compressions occurring along the depth axis are more challenging to discern. In this study, we propose a method to measure accurate intraoperative distances during MVD surgery using Depth-Anything-V2. By leveraging the optical properties of standard imaging equipment in conjunction with the depth estimation model, our method computes precise, absolute distances rather than relying solely on relative measurements, achieving distance estimation errors of less than 2 mm compared to intraoperative and preoperative reference measurements.

Finding refrigerants is challenging because they should be non-toxic, non-flammable, energy-saving and thermodynamically stable. In addition, they should have low global warming potentials to mitigate global warming. Traditionally, a large number of molecules could be studied by quantum chemical calculations or experimental measurements. However, these methods are often challenging to apply on a large scale due to their time-consuming and resource-intensive nature. In this aspect, we propose a fully data-driven screening with machine learning models to find promising refrigerant candidates. We used two prediction methods: a less accurate but explicable feedforward neural network and a more accurate graph neural network (GNN). Using GNN, non-toxic molecules were determined using positive-unlabeled learning from the molecular dataset, which partially contains toxicity information. Out of 33k molecules, 7 molecules were recommended as future refrigerants. Molecular properties of these molecules and their uncertainties were also reported using data-driven methods. These contributed to the identification of alternative refrigerants, which remains a challenging topic today. We believe that our approach can be used to search for non-toxic emerging molecules in the future.

This paper presents an LNA-embedded CT <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\Delta\Sigma$</tex> ADC for closed-loop neural recording. The frontend LNA is embedded in the loop filter of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\Delta\Sigma$</tex> ADC to achieve low input-referred noise (IRN) and high tolerance to stimulation artifacts. A 25-level feedback RDAC is realized with a 12-tap tri-level FIR-DAC and helps to linearize the LNA, resulting in high linearity over a wide input range. The FIR-DAC's delay is compensated by a novel feedforward compensation scheme to maintain loop stability. The DC-coupled LNA is chopped at a low frequency (~100kHz) and provides high input impedance, low offset, and low <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$1/f$</tex> noise without suffering from chopper artifacts. Implemented in a 65nm CMOS process, the ADC achieves an IRN of 62.5nV/√Hz, 86.7dB SNDR, 87.5dB DR, and 97.9dB SFDR, while consuming only <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$11.8\mu \mathrm{W}$</tex> in a 10kHz bandwidth. This corresponds to the state-of-the-art FoM of 176.8dB.

Traffic-related non-exhaust emissions, particularly from tire wear and brake wear, significantly contribute to urban road dust pollution and heavy metal accumulation. Despite growing concerns over tire-wear particles (TWP) as a source of microplastic pollution, the influence of road roughness (International Roughness Index, IRI), driving speed, and road slope on TWP generation and heavy-metal deposition remains underexplored. This study systematically investigated these relationships; field measurements were conducted across 57 road segments in Seoul, incorporating IRI assessments, road-dust sampling, and elemental analysis. Roads with IRI >3.0 m/km exhibited 11.9 % higher TWP concentrations than smoother roads did. When normalized by traffic volume (TWP/traffic), the strong correlation between IRI and per-vehicle TWP emissions confirmed that road-surface deterioration amplified tire wear independent of traffic density. Low-speed roads (<25 km/h) accumulated 24 % more TWP than high-speed roads (>40 km/h) did, suggesting that stop-and-go driving conditions enhance mechanical degradation. Downhill segments exhibited 0.2-2.1 % higher magnetic content than uphill segments did due to increased braking forces. This trend aligns with known brake-wear emissions, where Fe-, Cu-, and Zn-rich particles from brake pads and discs contribute significantly to on-road dust pollution. These findings provide empirical evidence linking road conditions, traffic behavior, and non-exhaust emissions, emphasizing the necessity of road maintenance and traffic-management strategies to mitigate TWP and heavy-metal deposition. The study offers valuable data for urban planners and policymakers in developing sustainable road infrastructure and emission-control measures, particularly in the context of Euro 7 regulations targeting non-exhaust emissions.

Abstract The complexity of modern battlefields demands advanced military training systems that prepare forces for realistic scenarios. Even though traditional training methods are effective, they are costly and time consuming and are associated with safety risks. Virtual training systems offer a safer and more cost‐effective alternative; however, current solutions often compromise realism because of the need for multiple sensors and wearable devices that can diminish immersion. To address these limitations, we propose a marker‐based, adaptive, virtual military training system (MAVMTS) that enhances realism by using a minimal set of fiducial markers and multiview cameras to estimate the trainee's posture and weapon orientation without cumbersome wearables. The system integrates action recognition to generate responsive virtual adversaries, thereby creating dynamic and immersive training environments. MAVMTS reduces considerably the equipment burden and enhances the realism of virtual military training, thereby offering a more effective solution for preparing personnel for modern warfare.

지상 네트워크는 제한된 커버리지로 인해 통신 사각지대와 장거리 통신에서 Quality of Service (QoS) 보장이 어렵다. 이를 해결하기 위해 저궤도 위성을 활용한 위성-지상 통합 네트워크 환경이 주목받고 있다. 저궤도 위성은 지상 네트워크와 먼 거리에 위치해 있음에도, 기존의 연구들은 전송 지연만 계산하고 위성의 긴 전파 거리를 고려하지 않아 적절한 라우팅 경로를 찾기 어렵다는 한계가 있다. 본 논문에서는 전송 및 전파 지연을 모두 고려한 weighted Expected Propagation and Transmission Time (wEPTT) 라우팅 메트릭과 utility 기반 QoS 라우팅 알고리즘을 제안한다. 실험 결과, 기존의 라우팅 메트릭에 비해 wEPTT는 낮은 지연 시간 및 높은 처리량을 갖는 것을 확인하였다.

Comparing different 3D shapes is challenging due to their irregularities. Motivated by the human visual system mechanism, where the entire 3D geometry is clearly perceived as a series of multiple projections, we propose a novel facial shape similarity measurement using multiview deep perceptual representations. We introduce a multiview disentangling scheme that accurately represents a facial mesh in multiple coordinates and the training strategy with view specificity and regional consistency to reliably train the network with multiple projections. View specificity pertains to the human visual perception to better recognize facial similarity. Regional consistency mitigates regional redundancy among views. Hence, robust perceptual features with respect to views are embedded and accurate similarity can be measured. Consequently, the view-specific integration scheme incorporates the similarities of all views, allowing for highly consistent measurement. The experiments demonstrate that the proposed similarity outperforms state-of-the-arts and significantly improves the details in terms of geometry and human perception.

Tire and road wear particles, major contributors to non-exhaust particulate matter emissions, are frequently resuspended into the atmosphere from road dust, posing significant environmental and health challenges. Conventional approaches to estimating emission factors (EFs) often rely on variables such as road dust loading, vehicle types, and road classifications; however, these methods typically neglect the critical influence of wind speed on resuspension dynamics. This study introduces a methodology that incorporates wind speed as a fundamental parameter to improve the accuracy of EF estimations for resuspended tire-wear particles (TWPs). Our approach utilizes particle size analysis, pyrolysis-gas chromatography-mass spectrometry for quantifying TWP content, and a wind-speed depended weighting factor (WFTWP) that accounts for the resuspension potential of particles. The average TWP content in road dust was determined to be 23,495 mg/kg (2.4 wt%), aligning with findings from previous urban studies. At a near-ground wind speed of 1.5 m/s, resuspended TWPs accounted for 2.6 % of the total resuspended dust mass, closely reflecting the original TWP proportion in road dust. Using modified EPA and Amato methods, calculated EF values ranged from 2.02 to 7.22 mg/vkm, with the Amato method's EF value (3.35 ± 2.21 mg/vkm) comparable to the EPA-derived EF for passenger cars (2.02 ± 0.55 mg/vkm) but showing significant variation for buses (7.22 ± 1.97 mg/vkm). Furthermore, the study found that as wind speed increased, the WFTWP also increased proportionally, directly impacting EF values. The results indicate the importance of incorporating wind dynamics into EF calculations to more accurately represent real-world resuspension behaviors. This methodology provides a practical tool for estimating the resuspension of TWPs under varying wind conditions and aids in refining emission inventories.

This article presents a continuous-time (CT) linear-exponential incremental ADC (IADC) that achieves 15-bit resolution at 250kHz bandwidth with 40 cycles for one conversion. It is based on an energy-efficient CT linear-exponential IADC, which alleviates the requirements of the power-hungry input buffer and loop filter. The proposed IADC employs a coarse 9-bit first-order IADC followed by a fine 8-bit cyclic ADC. The first-order IADC performs the coarse conversion by linearly accumulating input signals, resulting in a small thermal noise penalty. The residual quantization noise is exponentially reduced by the cyclic ADC, significantly shortening the conversion cycle. The cyclic ADC achieves the required accuracy by reconfiguring the loop filter of the coarse IADC and effectively compensating for the excessive loop delay. The prototype CT IADC is fabricated in a 65-nm CMOS process. With a 20MHz clock, it achieves 88.6-dB SNDR, 89.3-dB SNR, 90.1-dB DR, and 103.1-dB SFDR at a conversion rate of 500 kS/s. It consumes only 2.4 mW from a 1.2 V supply. It achieves the Schreier FoM (SNDR) of 168.8dB.

Purpose: This study aims to investigate and compare the incidence, demographic characteristics, clinical manifestations, preoperative diagnostic methods, anatomical classifications, associated anomalies, operative treatments, and postoperative outcomes of patients with intestinal atresia treated by the members of the Korean Association of Pediatric Surgeons (KAPS) through three nationwide surveys.Methods: KAPS conducted 3 national surveys in 1998, 2010, and 2024 to examine the patients diagnosed with intestinal atresia.In preparation for the survey, we developed a customized case registration form to obtain data on patient sex, birth weight, gestational age, clinical manifestations, preoperative diagnostic methods, anatomical types, associated anomalies, operative treatments, and postoperative outcomes.Authorized KAPS members completed the case registration form.Results: The first, second, and third national surveys included 218, 222, and 236 individuals diagnosed with intestinal atresia, respectively.The male-to-female ratios were 1.5:1, 1.1:1, and 1.1:1, respectively.The first, second, and third national surveys revealed that 34.3%, 43.3%, and 53.4% of patients were born before 37 weeks of gestation, respectively.Additionally, 28.7%, 32.0%, and 40.7% of patients had a birth weight under 2,500 g.In the third national survey, duodenoduodenostomy was the most common procedure, performed in 70 out of 82 patients diagnosed with duodenal atresia.Resection and anastomosis were the main surgical procedures conducted in 47 out of 54 cases of jejunal atresia and 74 out of 92 cases of ileal atresia.The mortality rates in the first, second, and third national surveys were 13.8%, 3.6%, and 1.3% respectively, with the lowest rate observed in the third national survey.Conclusion: These national surveys offer valuable insights into the current state of intestinal atresia, including specific surgical interventions and postoperative outcomes in South Korea.For pediatric surgeons aiming to enhance their understanding of intestinal atresia and its treatment options, these surveys could be an indispensable resource and guide.

Virtual reality (VR) makes it possible to provide immersive multimedia content composed of omnidirectional videos (ODVs). Towards enabling more immersive and satisfying VR content, methods are needed to manipulate VR scenes, taking into account perceptual factors related to viewers' quality of experience (QoE). For example, style transfer methods can be applied to VR content, allowing users to create artistic or surreal effects in their immersive environments. Here, we study perceptual factors that affect the sensation of stylized immersiveness, including color dynamics and spatio-temporal consistency. To do this, we introduce an immersiveness sensitivity model of luminance and color perception, and use it to measure the color dynamics and spatio-temporal consistency of stylized VR contents. We subsequently use this model to construct a perceptually-guided VR style transfer model called VR Style Transfer GAN (VRST-GAN). VRST-GAN learns to transfer a desired style into VR to enhance immersiveness by considering color dynamics while preserving spatio-temporal consistency. We demonstrate the effectiveness of VRST-GAN via qualitative and quantitative experiments. We also develop a VR Immersiveness Predictor (VR-IP) that is able to predict the sensation of immersiveness using the perceptual model. In our experiments, VR-IP predicts immersiveness with an accuracy of 91%.

When a phosphor film based on a photonic crystal (PhC) is excited at the photonic band-edge wavelength, the absorption of excitation light increases, which can potentially enhance the color-conversion efficiency. In this study, we modeled a two-dimensional (2D) PhC quantum dot (QD) film with a square-lattice structure using the finite-difference time-domain method to theoretically investigate its optical properties. The embedment of a thin-film layer with a high refractive index on the surface of the QD film enables an effective localization of excitation light within the phosphor. A numerical estimation shows that the optimized 2D PhC QD film can enhance the light absorption by up to 4.2 times with a monochromatic source and by up to 1.8 times with a broadband (FWHM~30 nm) source compared to a flat-type reference QD film.

With the development of computer vision technology, intelligent video surveillance systems have been developed for automatic monitoring. However, the problem of personal information protection has also emerged. Existing systems attempted to solve this problem by anonymizing a video by, for example, sending only low-dimensional abstract information such as a person’s 2D pose or blurring a person’s face in the video before sending it to the central cloud server. However, these approaches failed to balance scene-preservation and traffic efficiency, because abstract information is too limited for preserving the entire scene, and video modification generates massive traffic. This article proposes a novel intelligent video surveillance system to overcome such limitations that preserves the scene information and generates minimal traffic through video anonymization. The proposed system reconstructs 3D human models and estimates segmentation masks to preserve a scene captured by a surveillance camera in its entirety. Parametric models represent 3D human models with several sets of parameters, and dictionary coding compresses the segmentation mask with a high compression ratio. The system follows the edge-cloud architecture, where the edge node extracts and transmits the scene information and the central cloud server generates the final anonymized video. We demonstrate the effectiveness of the proposed system by conducting experiments on processing time, scene preservation, and traffic efficiency. Our proposed system runs in real-time ( \(&gt;\) 25fps) in a typical hardware setting and has a data compression ratio of more than 5,000 compared with raw data transfer while maintaining over 85% scene-preservation correlation with the original video.

외부 육각 볼트 머리 형태의 배플포머볼트에 대한 체적검사는 일반적으로 2개의 압전소자를 가지는 초음파 탐촉자가 lock-bar 양측의 볼트의 머리 부분에 접촉하고 볼트의 중심부로 향하는 6도의 굴절 종파를 발생시켜 수행된다. 이러한 탐촉자는 고정된 위치에서 한 각도의 초음파로 검사를 수행하기 때문에 검사 성능이 제한된다. 위상배열 초음파 기술 적용은 이러한 제한사항을 극복하고 배플포머볼트 초음파 검사 신뢰성을 향상시킬 수 있다. 하지만 기존의 상용 위상배열 초음파 장치에서는 탐촉자의 분리된 압전소자 그룹에 대해서는 lock-bar 하단의 위치한 결함 탐지에 유용한 피치캐치 모드 초음파 이미지 생성이 불가하다는 한계를 가지고 있다. 이번 연구에서는 배플포머볼트 검사를 위해 Full Matrix Capture (FMC)와 Total Focusing Method(TFM)에 기반한 위상배열 초음파 검사 기술을 개발하였다. 제안된 기법을 통해 피치캐치 모드에 대한 초음파 이미지를 합성할 수 있었고, 다양한 모드의 TFM 이미지는 배플포머볼트의 다양한 위치의 결함을 지시하고 평가함에 있어 유용하게 적용할 수 있음을 검증하였다.

The NEAMS Multiphysics Applications team aims at providing assessment of code useability and functionality for microreactor design and analyses, together with demonstration of their capabilities to properly capture the steady-state and time-dependent behavior of different microreactor concepts.In FY-2024, significant progress was achieved in improving multiphysics models of several microreactors systems: HP-MR, GC-MR and KRUSTY.These efforts focused on solving more complex multiphysics problems enabled by enhanced tools capability, verifying and validating results obtained, providing feedback to developers for suggested improvements, and sharing these models to facilitate user training.A series of new multiphysics transients were completed on the HP-MR (using Griffin/BISON/Sockeye) with core startup transient, control drum inadvertent rotation accident, and hydrogen leakage from hydride moderator (also including SWIFT).On the GC-MR, a new full-core model was developed and analyzed through a series of new multiphysics (Griffin/BISON/SAM) transients to simulate moderator leakage (also including SWIFT), flow blockage and coolant depressurization.Additional and updated TRISO failure analyses were completed on the HP-MR unit-cell and GC-MR assembly models leveraging improved TRISO modeling capabilities.The amount of SiC failure following accidental transients at end-of-life was null.However, GC-MR assembly TRISO analysis highlighted Pd penetration rate can be problematic and may require design changes on the studied microreactor concept.The neutronics discrepancies observed on the KRUSTY model in previous years were resolved using hybrid set of Monte Carlo/Deterministic cross-sections.The multiphysics (Griffin neutronics / BISON thermal-mechanics) 15 insertion transient simulation displayed good agreement when comparing with experimental data.Initial modeling of the 30 reactivity insertion also displays promising results.Such close agreement provides important validation data that can be leveraged by the NEAMS program and by microreactor vendors to support licensing of their technology.Finally, important experience was gathered with the NEAMS tools leading to several user feedback shared with tools developers, especially with regards to MOOSE mesh generator and Griffin.This project led to many publications demonstrating modeling capabilities, and to three models shared on the Virtual Test Bed.Assessment and validation of NEAMS tools for high-fidelity multiphysics transient modeling of microreactors

Most 360 virtual reality (VR) contents have been developed without considering that users could be affected by VR sickness. Accordingly, users' viewing safety has been steadily highlighted as a critical problem in the VR market. In this study, we investigate a novel VR sickness mitigation framework based on human visual characteristics for the rendered VR content. First, we build a large-scale 360 VR content database termed VRSP360 (VR Sickness and Presence 360) dedicated to the analysis of VR sickness and thoroughly conduct eye-tracking experiments to measure human perception. In the experiment, we observe that the users' gaze distribution is highly center-biased when they experience excessive VR sickness. From this observation, we design a foveated filtering framework that limits high-frequency textures in the peripheral view to mitigate VR sickness. Particularly, given the human visual system's (HVS) non-uniform resolution with respect to the fovea, we also adopt the foveation-based filtering method using the trade-off between sickness mitigation and presence conservation, which reduces any loss in perceptual quality despite the filtering. We further demonstrate that our framework can effectively compress visual information by applying foveated compression. In addition, we develop two metrics (visual texture index and perceptual information index) to measure the effective preservation of user-perceived information despite the filtration of peripheral vision textures by our proposed mitigation method. Through rigorous subjective evaluation on both original content and its VR-sickness-mitigated version, we demonstrate that the proposed framework successfully mitigates VR sickness with a reduction rate of $\sim$∼19% on the proposed dataset.

Despite acceleration in the use of 3D meshes, it is difficult to find effective mesh quality assessment algorithms that can produce predictions highly correlated with human subjective opinions. Defining mesh quality features is challenging due to the irregular topology of meshes, which are defined on vertices and triangles. To address this, we propose a novel 3D projective structural similarity index ( 3D- PSSIM) for meshes that is robust to differences in mesh topology. We address topological differences between meshes by introducing multi-view and multi-layer projections that can densely represent the mesh textures and geometrical shapes irrespective of mesh topology. It also addresses occlusion problems that occur during projection. We propose visual sensitivity weights that capture the perceptual sensitivity to the degree of mesh surface curvature. 3D- PSSIM computes perceptual quality predictions by aggregating quality-aware features that are computed in multiple projective spaces onto the mesh domain, rather than on 2D spaces. This allows 3D- PSSIM to determine which parts of a mesh surface are distorted by geometric or color impairments. Experimental results show that 3D- PSSIM can predict mesh quality with high correlation against human subjective judgments, across the presence of noise, even when there are large topological differences, outperforming existing mesh quality assessment models.

This paper presents the latest technology of displays that have recently been expanding from Rigid to a variety of form factors that can be modified. In a flexible display, different types of films are laminated to ensure panel reliability and to control stress distribution. However, the mechanism for how the characteristics of commercialized films affect the panel has not been properly identified, and it is not easy to adjust the physical properties to find key factors. So we formed a film with various characteristics using UV curing process and found key properties that can affect flexible OLED panel through process change. We were able to confirm the results of the experiment by analyzing and supplementing simulations based on physics. Finally, based on the mechanism proposed by comparing aspects between films, panel characteristics were verified and reliable.

This research delves into liquid fuel injection in scramjet combustors, with a focus on the effects of injector length-to-diameter ([Formula: see text]) ratios and temperatures, spanning from ambient to superheated states. Efficient atomization within the short residence times of liquid-fueled supersonic combustors is crucial for maximizing combustion efficiency. Initial investigations in a quiescent environment aimed to isolate the impact of injector geometry and temperature, utilizing quantitative analysis and diffused backlit illumination to demonstrate that heating fuel beyond its boiling point significantly enhances atomization and dispersion. A notable finding is that higher [Formula: see text] ratios led to a quicker initiation of flash boiling. Subsequent analyses under supersonic crossflow, using two-dimensional planar laser-induced fluorescence, support these observations. The study further employs spray structure and spatiotemporal characteristic analyses, coefficient of variation, and proper orthogonal decomposition, revealing enhanced flashing atomization at higher [Formula: see text] ratios. This underscores the critical role of injector design and temperature in optimizing fuel atomization in scramjet engines, highlighting the nuanced interplay between physical injector characteristics and thermal properties in achieving efficient combustion.

Existing emotional talking 3D facial animation primarily focus on animating emotional faces using a specific emotion condition. However, in real-world situations, no one consistently speaks with just one emotion. Thus, previous emotion-based approaches have very limited applicability in real-world applications. To address this issue, we propose SDETalk, a novel learning framework that animates the emotional talking faces by leveraging the emotional source from a speech. Unlike previous studies, which use static one-hot emotion conditions, the proposed network regresses complex emotional states from speech. It enables the network to animate natural facial animation from an emotional speech without using a specific emotional condition. Furthermore, we design the proposed method to produce head motions because head motion is an important factor to enhance the naturalness of talking face animation. By doing this, our approach simultaneously achieves accurate lip motion, natural expressions, and rhythmical head motions from emotional speech. Through extensive experiments in both qualitative and quantitative manners, it is demonstrated that our method outperforms other state-of-the-art methods by animating realistic and expressive 3D faces.

Denoising diffusion models have shown a powerful capacity for generating high-quality image samples by progressively removing noise. Inspired by this, we present a diffusion-based mesh denoiser that progressively removes noise from mesh. In general, the iterative algorithm of diffusion models attempts to manipulate the overall structure and fine details of target meshes simultaneously. For this reason, it is difficult to apply the diffusion process to a mesh denoising task that removes artifacts while maintaining a structure. To address this, we formulate a structure-preserving diffusion process. Instead of diffusing the mesh vertices to be distributed as zero-centered isotopic Gaussian distribution, we diffuse each vertex into a specific noise distribution, in which the entire structure can be preserved. In addition, we propose a topology-agnostic mesh diffusion model by projecting the vertex into multiple 2-D viewpoints to efficiently learn the diffusion using a deep network. This enables the proposed method to learn the diffusion of arbitrary meshes that have an irregular topology. Finally, the denoised mesh can be obtained via refinement based on 2-D projections obtained from reverse diffusion. Through extensive experiments, we demonstrate that our method outperforms the state-of-the-art mesh denoising methods in both quantitative and qualitative evaluations.

The increase in GPU-based AI applications, cloud-based gaming, and video streaming services has driven the need for new a graphics memory that operates at higher bandwidth and power efficiency than existing GDDR6 SDRAM, leading to the introduction of the GDDR7 standard [1]. Since performance degradation due to thermal throttling, power cost, and device reliability are major development considerations in high-power graphics applications, PAM3 signaling is applied on single-ended pins to improve bandwidth and power consumption, while maintaining the clock frequency [2]. However, new PAM3-related blocks supporting double the bandwidth inevitably increase in absolute power and temperature. In this paper, we present additional power reduction techniques, while maintaining SNR. The clocking architecture, with fast wake-up capabilities, can be partially disabled to provide active-standby current (IDD3N) as low as the power-down mode. The PAM3 TX and RX use a design approach that achieves high SNR and power efficiency.

In smart manufacturing systems (SMSs), flexible job-shop scheduling with transportation constraints (FJSPT) is essential to optimize solutions for maximizing productivity, considering production flexibility based on automated guided vehicles (AGVs). Recent developments in deep reinforcement learning (DRL)-based methods for FJSPT have encountered a scale generalization challenge. These methods underperform when applied to environment at scales different from their training set, resulting in low-quality solutions. To address this, we introduce a novel graph-based DRL method, named the Heterogeneous Graph Scheduler (HGS). Our method leverages locally extracted relational knowledge among operations, machines, and vehicle nodes for scheduling, with a graph-structured decision-making framework that reduces encoding complexity and enhances scale generalization. Our performance evaluation, conducted with benchmark datasets, reveals that the proposed method outperforms traditional dispatching rules, meta-heuristics, and existing DRL-based approaches in terms of makespan performance, even on large-scale instances that have not been experienced during training.

In recent years, the dance entertainment industry has experienced significant growth, driven by the desire of consumers to learn and improve their dancing skills. To effectively improve their skills, dancers require evaluation and feedback, which traditionally relies heavily on professional dancers. To address this challenge, researchers have proposed objective assessment methods for dance performance via kinematic data captured by sensors. However, these existing methods primarily focus on assessing the rhythmic accuracy of movements synchronized to music. In this paper, we propose Dance Quality Assessment (DanceQA) Framework to evaluate dance performance, considering choreographic factors that are important criteria in subjective DanceQA. We find that kinematic diversity and rhythmic alignment are significant choreographic factors from human perception perspective. Based on these factors, we design two metrics: kinematic information entropy (KIE) and kinematic-music beat similarity (BSIM). Our study demonstrates that these metrics are closely related to specific body parts in each choreography. To validate the effectiveness of our metrics, we capture dance performance by OptiTrack system providing precise three-dimensional data at very high sampling rate. We then label their dance quality via subjective test. The metrics give strong correlation with subjective opinion, but it is difficult to tell which body part is the most correlated. To comprehensively understand the dance quality, we propose choreographic quality transformers (CQTs), which learn the aforementioned choreographic factors by embedding KIE and BSIM into attention matrices. In numerous experiments, the CQTs outperforms previous methods, graph convolutional networks and multimodal transformers, at least by up to 0.146 in correlation coefficient.

Molecular core structures and R-groups are essential concepts in drug development. Integration of these concepts with conventional graph pre-training approaches can promote deeper understanding in molecules. We propose MolPLA, a novel pre-training framework that employs masked graph contrastive learning in understanding the underlying decomposable parts inmolecules that implicate their core structure and peripheral R-groups. Furthermore, we formulate an additional framework that grants MolPLA the ability to help chemists find replaceable R-groups in lead optimization scenarios. Experimental results on molecular property prediction show that MolPLA exhibits predictability comparable to current state-of-the-art models. Qualitative analysis implicate that MolPLA is capable of distinguishing core and R-group sub-structures, identifying decomposable regions in molecules and contributing to lead optimization scenarios by rationally suggesting R-group replacements given various query core templates. The code implementation for MolPLA and its pre-trained model checkpoint is available at https://github.com/dmis-lab/MolPLA

In this paper, we introduce a novel approach for refining depth estimation of multi-view stereo (MVS) applications by leveraging denoising diffusion models. Because recent state-of-the-art MVS methods rely on human heuristics to get a hypothesis of where the true depth lies, it is prone to produce erroneous and noisy results, particularly at the edge of the object or background. To ameliorate these issues, we present a plugin refinement process for depth estimate based on stochastic prior. Specifically, the stochastic distribution of depth maps is modeled by using the diffusion model. However, directly training depth maps on the diffusion model may introduce undesirable caveats such as holes. Thus, we use the knowledge distillation technique by using multi-view stereo output with its estimation confidence. Experimental results demonstrate the effectiveness of our proposed method compared to existing state-of-the-art MVS techniques.

In this paper, we introduce a robust 3D hand tracking framework using multi-view video, designed for the construction of a database. For hand pose estimation, real-world videos present frequent challenging scenarios, where current methods could fail, such as complex finger poses, two-hand interactions, and self-occluded situations. To address these challenges, our focus lies in enhancing the robustness and stability of 3D hand pose tracking. Firstly, we leverage multiple pre-trained 2D hand pose estimation algorithms, and combine them based on confidence and two-hand interactions. Additionally, to consider both cross-frame and cross-view consistency, our framework optimizes the entire joints with three objectives: data, guidance, and feature constraints. Throughout evaluation on real-world data, we show that the proposed method is robust to various challenging cases.