DR-NTU (Data) (Nanyang Technological University)
DR-NTU (Data) (Nanyang Technological University)Not a member yet
1955 research outputs found
Sort by
ClusteringSDF: Self-Organized Neural Implicit Surfaces for 3D Decomposition
No full text3D decomposition/segmentation remains a challenge as large-scale 3D annotated data is not readily available. Existing approaches typically leverage 2D machine-generated segments, integrating them to achieve 3D consistency. In this paper, we propose ClusteringSDF, a novel approach achieving both segmentation and reconstruction in 3D via the neural implicit surface representation, specifically the Signed Distance Function (SDF), where the segmentation rendering is directly integrated with the volume rendering of neural implicit surfaces. Although based on ObjectSDF++, ClusteringSDF no longer requires ground-truth segments for supervision while maintaining the capability of reconstructing individual object surfaces, relying purely on the noisy and inconsistent labels from pre-trained models. As the core of ClusteringSDF, we introduce a highly efficient clustering mechanism for lifting 2D labels to 3D. Experimental results on the challenging scenes from ScanNet and Replica datasets show that ClusteringSDF can achieve competitive performance compared to the state-of-the-art with significantly reduced training time
Related Data for: Enhancing cooperativity of molecular J-aggregates by resonantly coupled dielectric metasurfaces
No full textExperimental and numerical data for the paper "Enhancing cooperativity of molecular J-aggregates by resonantly coupled dielectric metasurfaces
Replication Data for: Optimization of Purcell-enhanced microcavities with the cylindrical finite-difference time-domain algorithm
No full textQuantum dots embedded in annular bullseye microcavities have been reported to be efficient single-photon sources. However, the proposed structures thus far often involve gratings of fixed periodicities, which may not be optimal considering the fact that Bessel functions are nonperiodic. In this paper, we present an optimization scheme for chirped annular microcavities where Purcell factors larger than 80 can be achieved through astute selection of the optimization function
Efficient Diffusion Model for Image Restoration by Residual Shifting
No full textWhile diffusion-based image restoration (IR) methods have achieved remarkable success, they are still limited by the low inference speed attributed to the necessity of executing hundreds or even thousands of sampling steps. Existing acceleration sampling techniques, though seeking to expedite the process, inevitably sacrifice performance to some extent, resulting in over-blurry restored outcomes. To address this issue, this study proposes a novel and efficient diffusion model for IR that significantly reduces the required number of diffusion steps. Our method avoids the need for post-acceleration during inference, thereby avoiding the associated performance deterioration. Specifically, our proposed method establishes a Markov chain that facilitates the transitions between the high-quality and low-quality images by shifting their residuals, substantially improving the transition efficiency. A carefully formulated noise schedule is devised to flexibly control the shifting speed and the noise strength during the diffusion process. Extensive experimental evaluations demonstrate that the proposed method achieves superior or comparable performance to current state-of-the-art methods on three classical IR tasks, namely image super-resolution, image inpainting, and blind face restoration, even only with four sampling steps
The benefit of soundscaping naturalistic sounds within urban residential areas: Mental fatigue recovery.
No full textPPG peak to peak raw data of participants
Related Data for: Intelligent Reflecting Surfaces-Assisted Hybrid THz/RF System Over Generalized Fading
No full textThis paper investigates a novel IRS-assisted hybrid framework that seamlessly combines both THz and radio frequency (RF) technologies using a selection combining (SC) scheme, thereby enhancing system reliability. The study incorporates the deterministic and statistical properties of IRS in both RF and THz domains employing a sophisticated spatial scattering chan-nel model across generalized α - µ fading channels. Specifically, the exact closed-form expressions for the probability density function (PDF) and cumulative distribution function (CDF) of the output signal-to-noise ratio (SNR) are derived for both THz and RF links. From this, the outage probability and average symbol error rate (SER) are derived
HAMSOM and ECOHAM model codes
No full textThe files contain the model codes and scripts used to create the data on which our analyses were based for the related publication. It is one file for the regional circulation model HAMSOM and another one for the regional ecosystem model ECOHAM
Data and Code for Influence of Language Dominance on Filipino’s Categorical Perception of English Phonemes
No full textDataset. This paper, seeks to investigate the phenomenon of categorical perception among Filipinos from both the Philippines and Singapore who speak English and a native language of the Philippines. The aim is to examine whether one’s dominant language being a language of the Philippines, as opposed to English, may interfere with their ability to accurately perceive some English phonemes as distinct categories. Repository contains materials for replication, as well as data and code for reproducing the results
MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoders
No full textMonocular 3D object detection aims for precise 3D localization and identification of objects from a single-view image. Despite its recent progress, it often struggles while handling pervasive object occlusions that tend to complicate and degrade the prediction of object dimensions, depths, and orientations. We design MonoMAE, a monocular 3D detector inspired by Masked Autoencoders that addresses the object occlusion issue by masking and reconstructing objects in the feature space. MonoMAE consists of two novel designs. The first is depth-aware masking that selectively masks certain parts of non-occluded object queries in the feature space for simulating occluded object queries for network training. It masks non-occluded object queries by balancing the masked and preserved query portions adaptively according to the depth information. The second is lightweight query completion that works with the depth-aware masking to learn to reconstruct and complete the masked object queries. With the proposed feature-space occlusion and completion, MonoMAE learns enriched 3D representations that achieve superior monocular 3D detection performance qualitatively and quantitatively for both occluded and non-occluded objects. Additionally, MonoMAE learns generalizable representations that can work well in new domains
