High-Frequency First: A Two-Stage Approach for Improving Image INR

• URL: http://arxiv.org/abs/2508.15582v2
• Date: Fri, 22 Aug 2025 08:30:16 GMT
• Title: High-Frequency First: A Two-Stage Approach for Improving Image INR
• Authors: Sumit Kumar Dam, Mrityunjoy Gain, Eui-Nam Huh, Choong Seon Hong,
• Abstract summary: Implicit Neural Representations (INRs) have emerged as a powerful alternative to traditional pixel-based formats.<n>A key challenge lies in the spectral bias of neural networks, which tend to favor low-frequency components while struggling to capture high-frequency details.<n>We introduce a two-stage training strategy where a neighbor-aware soft mask adaptively assigns higher weights to pixels with strong local variations.
• Score: 13.070432644808806
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Implicit Neural Representations (INRs) have emerged as a powerful alternative to traditional pixel-based formats by modeling images as continuous functions over spatial coordinates. A key challenge, however, lies in the spectral bias of neural networks, which tend to favor low-frequency components while struggling to capture high-frequency (HF) details such as sharp edges and fine textures. While prior approaches have addressed this limitation through architectural modifications or specialized activation functions, we propose an orthogonal direction by directly guiding the training process. Specifically, we introduce a two-stage training strategy where a neighbor-aware soft mask adaptively assigns higher weights to pixels with strong local variations, encouraging early focus on fine details. The model then transitions to full-image training. Experimental results show that our approach consistently improves reconstruction quality and complements existing INR methods. As a pioneering attempt to assign frequency-aware importance to pixels in image INR, our work offers a new avenue for mitigating the spectral bias problem.

Related papers

• Learning Multi-scale Spatial-frequency Features for Image Denoising [58.883244886588336]
  We propose a novel multi-scale adaptive dual-domain network (MADNet) for image denoising.<n>We use image pyramid inputs to restore noise-free results from low-resolution images.<n>In order to realize the interaction of high-frequency and low-frequency information, we design an adaptive spatial-frequency learning unit.
  arXiv  Detail & Related papers  (2025-06-19T13:28:09Z)
• FADPNet: Frequency-Aware Dual-Path Network for Face Super-Resolution [70.61549422952193]
  Face super-resolution (FSR) under limited computational costs remains an open problem.<n>Existing approaches typically treat all facial pixels equally, resulting in suboptimal allocation of computational resources.<n>We propose FADPNet, a Frequency-Aware Dual-Path Network that decomposes facial features into low- and high-frequency components.
  arXiv  Detail & Related papers  (2025-06-17T02:33:42Z)
• Super-temporal-resolution Photoacoustic Imaging with Dynamic Reconstruction through Implicit Neural Representation in Sparse-view [4.333674832664625]
  Implicit Neural Representation (INR) has emerged as a powerful deep learning tool for solving inverse problems with sparse data.<n>In this work, we propose an INR-based method to improve dynamic photoacoustic image reconstruction from sparse-views.<n>The proposed INR represents dynamic photoacoustic images as implicit functions and encodes them into a neural network.
  arXiv  Detail & Related papers  (2025-05-29T06:36:44Z)
• Unified Image Restoration and Enhancement: Degradation Calibrated Cycle Reconstruction Diffusion Model [8.713784455593778]
  CycleRDM is a novel framework designed to unify restoration and enhancement tasks.<n>It learns the mapping relationships among the degraded domain, the rough normal domain, and the normal domain.<n>To improve restoration quality, we design a feature gain module for the decomposed wavelet high-frequency domain.
  arXiv  Detail & Related papers  (2024-12-19T08:33:33Z)
• LeRF: Learning Resampling Function for Adaptive and Efficient Image Interpolation [64.34935748707673]
  Recent deep neural networks (DNNs) have made impressive progress in performance by introducing learned data priors. We propose a novel method of Learning Resampling (termed LeRF) which takes advantage of both the structural priors learned by DNNs and the locally continuous assumption. LeRF assigns spatially varying resampling functions to input image pixels and learns to predict the shapes of these resampling functions with a neural network.
  arXiv  Detail & Related papers  (2024-07-13T16:09:45Z)
• Enhanced Low-Dose CT Image Reconstruction by Domain and Task Shifting Gaussian Denoisers [3.4748713192043876]
  Computed tomography from a low radiation dose (LDCT) is challenging due to high noise in the projection data.<n>We propose a method combining the simplicity and efficiency of two-stage methods with state-of-the-art reconstruction quality.
  arXiv  Detail & Related papers  (2024-03-06T08:51:09Z)
• Informative Rays Selection for Few-Shot Neural Radiance Fields [0.3599866690398789]
  KeyNeRF is a simple yet effective method for training NeRF in few-shot scenarios by focusing on key informative rays. Our approach performs favorably against state-of-the-art methods, while requiring minimal changes to existing NeRFs.
  arXiv  Detail & Related papers  (2023-12-29T11:08:19Z)
• Distance Weighted Trans Network for Image Completion [52.318730994423106]
  We propose a new architecture that relies on Distance-based Weighted Transformer (DWT) to better understand the relationships between an image's components. CNNs are used to augment the local texture information of coarse priors. DWT blocks are used to recover certain coarse textures and coherent visual structures.
  arXiv  Detail & Related papers  (2023-10-11T12:46:11Z)
• Rank-Enhanced Low-Dimensional Convolution Set for Hyperspectral Image Denoising [50.039949798156826]
  This paper tackles the challenging problem of hyperspectral (HS) image denoising. We propose rank-enhanced low-dimensional convolution set (Re-ConvSet) We then incorporate Re-ConvSet into the widely-used U-Net architecture to construct an HS image denoising method.
  arXiv  Detail & Related papers  (2022-07-09T13:35:12Z)
• Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline [100.5353614588565]
  We propose to incorporate the domain knowledge of the LDR image formation pipeline into our model. We model the HDRto-LDR image formation pipeline as the (1) dynamic range clipping, (2) non-linear mapping from a camera response function, and (3) quantization. We demonstrate that the proposed method performs favorably against state-of-the-art single-image HDR reconstruction algorithms.
  arXiv  Detail & Related papers  (2020-04-02T17:59:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.