neural field
PulseAugur coverage of neural field — every cluster mentioning neural field across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New framework enhances image compression for machine and human vision
Researchers have developed a new framework called IGDA-ICMH for image compression, specifically designed to improve performance for both machine and human vision tasks. This framework addresses the issue of unequal cont…
-
TensorSSM uses dynamic state space models for tensor compression
Researchers have introduced TensorSSM, a novel Implicit Neural Representation (INR) designed for functional tensor decomposition. This method utilizes a dynamic State Space Model (SSM) to memorize spatial positions with…
-
HEALPix-based compression improves omnidirectional image quality
Researchers have developed a novel method for compressing omnidirectional images using a HEALPix-based implicit neural representation. This approach addresses limitations of existing equirectangular projection methods b…
-
AcoustiPrint: Physics-Guided Neural Design for Manufacturable Acoustic Structures
Researchers have developed AcoustiPrint, a novel physics-informed neural network designed to create manufacturable 3D acoustic structures. This system addresses the challenge of integrating non-differentiable fabricatio…
-
Neural networks compared for indoor environment prediction
A new research paper compares two types of geometry-conditioned neural networks, Canvas-based neural networks (CaNN) and coordinate-based neural networks (CoNN), for predicting indoor airflow and temperature. The study …
-
New LiDAR method reconstructs hidden scenes through transparent occluders
Researchers have developed a new framework to reconstruct 3D scenes from single-photon LiDAR data, even when partially obscured by transparent materials. This method addresses the challenge of mixed returns from both fo…
-
New neural representation method boosts hyperspectral video compression
Researchers have developed a novel method for compressing hyperspectral video using implicit neural representations. This new approach significantly outperforms traditional hyperspectral image compression techniques, ac…
-
JSolver framework enables multi-material decomposition from single-energy CT scans
Researchers have developed JSolver, a novel framework for multi-material decomposition (MMD) using single-energy CT (SECT) projections. Unlike traditional methods that require spectral CT scanners, JSolver jointly recon…
-
New blockwise optimizer enhances cubic Newton methods for large-scale neural networks
This paper introduces a novel blockwise optimizer designed to improve the efficiency of cubic regularized Newton methods for large-scale neural network training. The proposed method handles arbitrarily large parameter t…
-
New neural representation method enhances 3D gravity inversion
Researchers have developed a novel unsupervised method for 3D gravity inversion using depth-aware implicit neural representations. This approach represents the subsurface density volume with multiple neural networks opt…
-
New neural representation method improves longitudinal MRI analysis
Researchers have developed a novel patient-specific conditional implicit neural representation (INR) designed to handle missing sequences and varying resolutions in longitudinal multiparametric MRI data. This model trea…
-
New SIINR framework enhances clinical dMRI resolution with uncertainty quantification
Researchers have developed SIINR, a novel framework for enhancing the resolution of diffusion Magnetic Resonance Imaging (dMRI) data. This method not only improves structural detail in clinical dMRI but also quantifies …
-
New AI framework reconstructs fetal brain MRI with enhanced accuracy
Researchers have developed PRIME-SVR, a novel implicit neural representation framework for reconstructing high-resolution 3D fetal brain volumes from 2D MRI slice stacks. This method is the first to enable quantitative …
-
New LIIFusion framework boosts generative MEF efficiency and quality
Researchers have introduced LIIFusion, a novel coarse-to-fine framework designed to improve the efficiency and quality of generative Multi-Exposure Fusion (MEF). This approach addresses the computational expense and str…
-
New CHOIR method enhances implicit neural representations for multi-dimensional data
Researchers have introduced a new method called Calibrated Harmonic Overlaid Implicit Neural Representation (CHOIR) to improve the stability and performance of implicit neural representations (INRs) for multi-dimensiona…
-
PatchINR reduces INR inference latency by 75% using patch-based approach
Researchers have developed PatchINR, a novel patch-based approach to Implicit Neural Representations (INRs) that significantly reduces computational costs for high-resolution signal modeling. By processing non-overlappi…
-
Neural fields improve weather prediction data assimilation
Researchers have developed a novel neural field-based approach to Four-Dimensional Variational Data Assimilation (4DVAR), a critical but computationally intensive process in numerical weather prediction. This new method…
-
PiG-Avatar generates realistic 3D avatars with neural fields
Researchers have introduced PiG-Avatar, a novel method for generating realistic 3D avatars. This approach decouples avatar geometry from body template surfaces, allowing for more accurate representation of complex cloth…