DR-NTU (Data) (Nanyang Technological University)
DR-NTU (Data) (Nanyang Technological University)Not a member yet
1955 research outputs found
Sort by
Related Data for: Strong Collective Responses in Quasi-1D Metadevices for Millimeter Electronics
No full textThis dataset contains MATLAB code used to generate key simulation results for the IEEE Journal of Microwaves manuscript on electronic metadevices. Specifically, the scripts calculate and plot the normalized contact resistance (Rc) as a function of (1) the number of stripes (N) for different stripe lengths, and (2) stripe length (L) for different values of N. The code performs theoretical calculations using finite element method (FEM)
Learning 3D Garment Animation from Trajectories of A Piece of Cloth
No full textGarment animation is ubiquitous in various applications, such as virtual reality, gaming, and film producing. Recently, learning-based approaches obtain compelling performance in animating diverse garments under versatile scenarios. Nevertheless, to mimic the deformations of the observed garments, data-driven methods require large scale of garment data, which are both resource-wise expensive and time-consuming. In addition, forcing models to match the dynamics of observed garment animation may hinder the potentials to generalize to unseen cases. In this paper, instead of using garment-wise supervised-learning we adopt a disentangled scheme to learn how to animate observed garments: 1). learning constitutive behaviors from the observed cloth; 2). dynamically animate various garments constrained by the learned constitutive laws. Specifically, we propose Energy Unit network (EUNet) to model the constitutive relations in the format of energy. Without the priors from analytical physics models and differentiable simulation engines, EUNet is able to directly capture the constitutive behaviors from the observed piece of cloth and uniformly describes the change of energy caused by deformations, such as stretching and bending. We further apply the pre-trained EUNet to animate various garments based on energy optimizations. The disentangled scheme alleviates the need of garment data and enables us to utilize the dynamics of a piece of cloth for animating garments. Experiments show that while EUNet effectively delivers the energy gradients due to the deformations, models constrained by EUNet achieve more stable and physically plausible performance comparing with those trained in garment-wise supervised manner
TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment
No full textMultivariate time series forecasting (MTSF) aims to learn temporal dynamics among variables to forecast future time series. Existing statistical and deep learning-based methods suffer from limited learnable parameters and small-scale training data. Recently, large language models (LLMs) combining time series with textual prompts have achieved promising performance in MTSF. However, we discovered that current LLM-based solutions fall short in learning disentangled embeddings. We introduce TimeCMA, an intuitive yet effective framework for MTSF via cross-modality alignment. Specifically, we present a dual-modality encoding with two branches: the time series encoding branch extracts disentangled yet weak time series embeddings, and the LLM-empowered encoding branch wraps the same time series with text as prompts to obtain entangled yet robust prompt embeddings. As a result, such a cross-modality alignment retrieves both disentangled and robust time series embeddings, ``the best of two worlds'', from the prompt embeddings based on time series and prompt modality similarities. As another key design, to reduce the computational costs from time series with their length textual prompts, we design an effective prompt to encourage the most essential temporal information to be encapsulated in the last token: only the last token is passed to downstream prediction. We further store the last token embeddings to accelerate inference speed. Extensive experiments on eight real datasets demonstrate that TimeCMA outperforms state-of-the-arts
MatAnyone: Stable Video Matting with Consistent Memory Propagation
No full textAuxiliary-free human video matting methods, which rely solely on input frames, often struggle with complex or ambiguous backgrounds. To tackle this, we propose MatAnyone, a practical framework designed for target-assigned video matting. Specifically, building on a memory-based framework, we introduce a consistent memory propagation module via region-adaptive memory fusion, which adaptively combines memory from the previous frame. This ensures stable semantic consistency in core regions while maintaining fine details along object boundaries. For robust training, we present a larger, high-quality, and diverse dataset for video matting. Additionally, we incorporate a novel training strategy that efficiently leverages large-scale segmentation data, further improving matting stability. With this new network design, dataset, and training strategy, MatAnyone delivers robust, accurate video matting in diverse real-world scenarios, outperforming existing methods. The code and model will be publicly available
Preregistration Documents for "Causal language in child-directed speech in Singapore"
No full textCausal reasoning is an important cognitive competency that allows us to make predictions, categorise items, make decisions, problem-solve and more (Waldmann & Hagmayer, 2013). A causal event structure involves a cause: a prior event that occurs, and an effect: a result that occurs because of the prior event. When describing this cause-effect relationship, people use causal language to describe the events. Causal language could include phrases like “as a result”, or sentence structures like “because… so…”. Human depth of causal understanding seems to develop from a young age. Infants start using and learning causal language from 12-24 months (Gopnik, 1982). Research shows that structural cues of causal language facilitates casual understanding in young children, and parental use of causal language can predict children’s causal verb comprehension (Aktan-Erciyes & Göksun, 2021; Ger et al., 2021). Given the importance of causal reasoning and the influence of causal language on understanding on causality, this study aims to investigate the developmental trajectory of causal language used by parents with their children in the Singaporean context
Related Data for: High-efficiency Ultraviolet Generation in Resonance-Free Antiresonant Hollow-core Fiber
No full textThe data for the manuscript "High-efficiency Ultraviolet Generation in Resonance-Free Antiresonant Hollow-core Fiber
Proactive career planning and development for Singapore Female Health Professionals
No full textFindings from our previous study on Singapore female health professionals’ (SFHPs) career decisions showed that SHFPs often take a passive approach to their career planning and development, and subsequently face challenges in their careers. They often have jobs and careers incompatible with their life aspirations and needs, poor levels of job and career satisfaction, and weak work engagement.
This study investigates the career planning and development behaviours of SFHPs, designing a programme to help increase their proactivity in lifelong career planning, taking into account their personal needs and life plans. Subsequently we will construct an assessment tool to evaluate effectiveness of career planning and development initiatives for SFHPs. This mixed-methods study will be conducted in three phases: (1) developing a model describing SFHPs’ career management behaviour from a representative quantitative survey and qualitative interviews; (2) designing and piloting a programme encouraging proactive career planning and development in SFHPs through co-design workshops with various stakeholders (healthcare employers, human resources professionals, and SFHPs); and (3) constructing an assessment tool to evaluate the effectiveness of career management programmes based on data from previous study phases and a scoping review of existing literature. This assessment tool will be validated using confirmatory factor analysis
Scale-Dependent Inverse Temperature Features Associated with Crashes in the US and Japanese Stock Markets
No full textPaper accepted by the Complexity Journal
Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration
No full textAlthough learning-based image restoration methods have made significant progress, they still struggle with limited generalization to real-world scenarios due to the substantial domain gap caused by training on synthetic data. Existing methods address this issue by improving data synthesis pipelines, estimating degradation kernels, employing deep internal learning, and performing domain adaptation and regularization. Previous domain adaptation methods have sought to bridge the domain gap by learning domain-invariant knowledge in either feature or pixel space. However, these techniques often struggle to extend to low-level vision tasks within a stable and compact framework. In this paper, we show that it is possible to perform domain adaptation via the noise space using diffusion models. In particular, by leveraging the unique property of how auxiliary conditional inputs influence the multi-step denoising process, we derive a meaningful diffusion loss that guides the restoration model in progressively aligning both restored synthetic and real-world outputs with a target clean distribution. We refer to this method as denoising as adaptation. To prevent shortcuts during joint training, we present crucial strategies such as channel-shuffling layer and residual-swapping contrastive learning in the diffusion model. They implicitly blur the boundaries between conditioned synthetic and real data and prevent the reliance of the model on easily distinguishable features. Experimental results on three classical image restoration tasks, namely denoising, deblurring, and deraining, demonstrate the effectiveness of the proposed method
Replication Data for: On-chip active supercoupled topological cavity
No full textOn-chip photonic resonant cavity plays a critical role in widespread applications including lasing, sensing, and spectroscopy. However, the excitation of these cavities typically relies on evanescent coupling within sub-wavelength distances, limiting flexible and precise chip integration. Here, we demonstrate an on-chip supercoupled topological cavity which is critically coupled at 2.3-wavelength distance from a bus waveguide and remains excited even at 3.2 wavelengths, based on the supercoupling mechanism enabled by the valley vortex flow. Optothermal heating facilitates tunable quality factors and dynamic control of the supercoupling condition, allowing transitions from overcoupling to undercoupling through the critical point. Our discovery extends the waveguide-cavity excitation distance to multiple wavelengths, unlocking new possibilities for designing and controlling on-chip resonant devices, including supercoupled lasers, sensors, and modulators