VSANet: View-aware Sparse Attention Network for Light Field Image Denoising

• URL: http://arxiv.org/abs/2606.24737v1
• Date: Tue, 23 Jun 2026 16:00:44 GMT
• Title: VSANet: View-aware Sparse Attention Network for Light Field Image Denoising
• Abstract summary: VSANet is a view-aware sparse attention network for light field (LF) image denoising.<n>We show VSANet outperforms state-of-the-art LF denoising methods.
• Score: 15.000821535291742
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Light field (LF) image denoising is challenging due to the high-dimensional structure of LF data. While noise is independent across sub-aperture images, scene content exhibits strong cross-view correlations. We introduce VSANet, a view-aware sparse attention network for LF denoising. Specifically, we propose a view-aware sparse attention (VSA) block that represents the 4D LF feature map as a unified spatial-angular token space and performs cross-view aggregation via locality-sensitive hashing-based sparse attention. This enables global feature interactions with linear complexity, effectively exploiting LF correlations across views and spatial locations. In addition, we design a feature refinement (FR) block to emphasize informative features in spatial, angular, and epipolar subspaces. The VSA and FR blocks are integrated within a sequential attention refinement module, forming the core of VSANet. Experiments demonstrate VSANet outperforms stateof-the-art LF denoising methods.

Related papers

• Non-Learning Low-Light Stereo Vision [12.842377848287045]
  We present a non-learning framework for disparity estimation from severely noisy images.<n>Using the Field of Junctions (FoJ), it retains coarse visual features stable under severe noise for cost volume construction while discarding fine textures inseparable from photon noise.<n>The resulting structural information guides boundary-aware Semi-Global Matching (SGM) that dynamically adapts smoothness penalties to preserve true disparity discontinuities.
  arXiv  Detail & Related papers  (2026-05-29T21:43:38Z)
• LFIC-DRASC: Deep Light Field Image Compression Using Disentangled Representation and Asymmetrical Strip Convolution [51.909036244222904]
  We propose an end-to-end deep LF Image Compression method using Disentangled Representation and Asymmetrical Strip Convolution. Experimental results demonstrate that the proposed LFIC-DRASC achieves an average of 20.5% bit rate reductions.
  arXiv  Detail & Related papers  (2024-09-18T05:33:42Z)
• Rethinking Transformer-Based Blind-Spot Network for Self-Supervised Image Denoising [94.09442506816724]
  Blind-spot networks (BSN) have been prevalent neural architectures in self-supervised image denoising (SSID)<n>We build a Transformer-based Blind-Spot Network (TBSN) which shows strong local fitting and global perspective abilities.
  arXiv  Detail & Related papers  (2024-04-11T15:39:10Z)
• Hybrid Convolutional and Attention Network for Hyperspectral Image Denoising [54.110544509099526]
  Hyperspectral image (HSI) denoising is critical for the effective analysis and interpretation of hyperspectral data. We propose a hybrid convolution and attention network (HCANet) to enhance HSI denoising. Experimental results on mainstream HSI datasets demonstrate the rationality and effectiveness of the proposed HCANet.
  arXiv  Detail & Related papers  (2024-03-15T07:18:43Z)
• Diffusion-based Light Field Synthesis [50.24624071354433]
  LFdiff is a diffusion-based generative framework tailored for LF synthesis. We propose DistgUnet, a disentanglement-based noise estimation network, to harness comprehensive LF representations. Extensive experiments demonstrate that LFdiff excels in synthesizing visually pleasing and disparity-controllable light fields.
  arXiv  Detail & Related papers  (2024-02-01T13:13:16Z)
• Spectral Enhanced Rectangle Transformer for Hyperspectral Image Denoising [64.11157141177208]
  We propose a spectral enhanced rectangle Transformer to model the spatial and spectral correlation in hyperspectral images. For the former, we exploit the rectangle self-attention horizontally and vertically to capture the non-local similarity in the spatial domain. For the latter, we design a spectral enhancement module that is capable of extracting global underlying low-rank property of spatial-spectral cubes to suppress noise.
  arXiv  Detail & Related papers  (2023-04-03T09:42:13Z)
• Spatially Adaptive Self-Supervised Learning for Real-World Image Denoising [73.71324390085714]
  We propose a novel perspective to solve the problem of real-world sRGB image denoising. We take into account the respective characteristics of flat and textured regions in noisy images, and construct supervisions for them separately. We present a locally aware network (LAN) to meet the requirement, while LAN itself is supervised with the output of BNN.
  arXiv  Detail & Related papers  (2023-03-27T06:18:20Z)
• Learning Non-Local Spatial-Angular Correlation for Light Field Image Super-Resolution [36.69391399634076]
  Exploiting spatial-angular correlation is crucial to light field (LF) image super-resolution (SR) We propose a simple yet effective method to learn the non-local spatial-angular correlation for LF image SR. Our method can fully incorporate the information from all angular views while achieving a global receptive field along the epipolar line.
  arXiv  Detail & Related papers  (2023-02-16T03:40:40Z)
• I See-Through You: A Framework for Removing Foreground Occlusion in Both Sparse and Dense Light Field Images [25.21481624956202]
  A light field (LF) camera captures rich information from a scene. Using the information, the LF de-occlusion (LF-DeOcc) task aims to reconstruct the occlusion-free center view image. We propose a framework, ISTY, which is defined and divided into three roles: (1) extract LF features, (2) define the occlusion, and (3) inpaint occluded regions. In experiments, qualitative and quantitative results show that the proposed framework outperforms state-of-the-art LF-DeOcc methods in both sparse and dense LF datasets.
  arXiv  Detail & Related papers  (2023-01-16T12:25:42Z)
• SSCAN: A Spatial-spectral Cross Attention Network for Hyperspectral Image Denoising [12.873607414761093]
  We propose a novel HSI denoising network, termed SSCAN, that combines group convolutions and attention modules. We propose a spectral-spatial attention block (SSAB) to exploit the spatial and spectral information in hyperspectral images in an effective manner. The experimental results indicate that the proposed SSCAN outperforms several state-of-the-art HSI denoising algorithms.
  arXiv  Detail & Related papers  (2021-05-23T14:36:17Z)

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.