Less is More: Skim Transformer for Light Field Image Super-resolution

• URL: http://arxiv.org/abs/2407.15329v2
• Date: Sun, 10 Aug 2025 03:42:07 GMT
• Title: Less is More: Skim Transformer for Light Field Image Super-resolution
• Authors: Zeke Zexi Hu, Haodong Chen, Hui Ye, Xiaoming Chen, Vera Yuk Ying Chung, Yiran Shen, Weidong Cai,
• Abstract summary: We present SkimLFSR, an efficient yet powerful network for light field super-resolution (LFSR)<n>SkimLFSR achieves state-of-the-art results surpassing the best existing method by an average of 0.59 dB and 0.35 dB in PSNR at the 2x and 4x tasks, respectively.<n>These findings highlight its effectiveness and adaptability as a promising paradigm for light field image processing.
• Score: 13.565016383731697
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: A light field image captures scenes through an array of micro-lenses, providing a rich representation that encompasses spatial and angular information. While this richness comes at the cost of significant data redundancy, most existing light field methods still tend to indiscriminately utilize all the information from sub-aperture images (SAIs) in an attempt to harness every visual cue regardless of their disparity significance. However, this paradigm inevitably leads to disparity entanglement, a fundamental cause of inefficiency in light field image processing. To address this limitation, we introduce the Skim Transformer, a novel architecture inspired by the ``less is more" philosophy. Unlike conventional light field Transformers, our Skim Transformer features a multi-branch structure where each branch is dedicated to a specific disparity range by constructing its attention score matrix over a skimmed subset of SAIs, rather than all of them. Building upon this core component, we present SkimLFSR, an efficient yet powerful network for light field super-resolution (LFSR). Requiring only 67\% of parameters, SkimLFSR achieves state-of-the-art results surpassing the best existing method by an average of 0.59 dB and 0.35 dB in PSNR at the 2x and 4x tasks, respectively. Through in-depth analyses, we reveal that SkimLFSR, guided by the predefined skimmed SAI sets as prior knowledge, demonstrates distinct disparity-aware behaviors in attending to visual cues. These findings highlight its effectiveness and adaptability as a promising paradigm for light field image processing.

Related papers

• Compressive Imaging Reconstruction via Tensor Decomposed Multi-Resolution Grid Encoding [50.54887630778593]
  Compressive imaging (CI) reconstruction aims to recover high-dimensional images from low-dimensional measurements compressed.<n>Existing unsupervised representations may struggle to achieve a desired balance between representation ability and efficiency.<n>We propose Decomposed multi-resolution Grid encoding (GridTD), an unsupervised continuous representation framework for CI reconstruction.
  arXiv  Detail & Related papers  (2025-07-10T12:36:20Z)
• DFVO: Learning Darkness-free Visible and Infrared Image Disentanglement and Fusion All at Once [57.15043822199561]
  A Darkness-Free network is proposed to handle Visible and infrared image disentanglement and fusion all at Once (DFVO)<n>DFVO employs a cascaded multi-task approach to replace the traditional two-stage cascaded training (enhancement and fusion)<n>Our proposed approach outperforms state-of-the-art alternatives in terms of qualitative and quantitative evaluations.
  arXiv  Detail & Related papers  (2025-05-07T15:59:45Z)
• QDM: Quadtree-Based Region-Adaptive Sparse Diffusion Models for Efficient Image Super-Resolution [54.67891514843853]
  We propose the Quadtree Diffusion Model (QDM), a region-adaptive diffusion framework. By guiding the diffusion with a quadtree derived from the low-quality input, QDM identifies key regions-represented by leaf nodes-where fine detail is essential. Experiments demonstrate QDM's effectiveness in high-resolution SR tasks across diverse image types, particularly in medical imaging.
  arXiv  Detail & Related papers  (2025-03-15T06:50:30Z)
• MLI-NeRF: Multi-Light Intrinsic-Aware Neural Radiance Fields [21.057216351934688]
  Current methods for extracting intrinsic image components, such as reflectance and shading, rely on statistical priors. We propose MLI-NeRF, which integrates textbfMultiple textbfLight information in textbfIntrinsic-aware textbfNeural textbfRadiance textbfFields.
  arXiv  Detail & Related papers  (2024-11-26T08:57:38Z)
• Effective Diffusion Transformer Architecture for Image Super-Resolution [63.254644431016345]
  We design an effective diffusion transformer for image super-resolution (DiT-SR) In practice, DiT-SR leverages an overall U-shaped architecture, and adopts a uniform isotropic design for all the transformer blocks. We analyze the limitation of the widely used AdaLN, and present a frequency-adaptive time-step conditioning module.
  arXiv  Detail & Related papers  (2024-09-29T07:14:16Z)
• LGFN: Lightweight Light Field Image Super-Resolution using Local Convolution Modulation and Global Attention Feature Extraction [5.461017270708014]
  We propose a lightweight model named LGFN which integrates the local and global features of different views and the features of different channels for LF image SR. Our model has a parameter of 0.45M and a FLOPs of 19.33G which has achieved a competitive effect.
  arXiv  Detail & Related papers  (2024-09-26T11:53:25Z)
• Dual-Domain Deep D-bar Method for Solving Electrical Impedance Tomography [5.112764609048122]
  The regularized D-bar method is one of the most prominent methods for solving Electrical Impedance Tomography (EIT) problems. D-bar images often present low contrast and low resolution due to the absence of accurate high-frequency information. We propose a dual-domain neural network architecture to retrieve high-contrast D-bar image sequences from low-contrast D-bar images.
  arXiv  Detail & Related papers  (2024-05-12T21:55:02Z)
• Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring [25.36888929483233]
  We propose a multi-scale network based on single-input and multiple-outputs(SIMO) for motion deblurring. We combine the characteristics of real-world trajectories with a learnable wavelet transform module to focus on the directional continuity and frequency features of the step-by-step transitions between blurred images to sharp images.
  arXiv  Detail & Related papers  (2023-12-29T02:59:40Z)
• Low-Light Image Enhancement with Illumination-Aware Gamma Correction and Complete Image Modelling Network [69.96295927854042]
  Low-light environments usually lead to less informative large-scale dark areas. We propose to integrate the effectiveness of gamma correction with the strong modelling capacities of deep networks. Because exponential operation introduces high computational complexity, we propose to use Taylor Series to approximate gamma correction.
  arXiv  Detail & Related papers  (2023-08-16T08:46:51Z)
• Enhancing Low-light Light Field Images with A Deep Compensation Unfolding Network [52.77569396659629]
  This paper presents the deep compensation network unfolding (DCUNet) for restoring light field (LF) images captured under low-light conditions. The framework uses the intermediate enhanced result to estimate the illumination map, which is then employed in the unfolding process to produce a new enhanced result. To properly leverage the unique characteristics of LF images, this paper proposes a pseudo-explicit feature interaction module.
  arXiv  Detail & Related papers  (2023-08-10T07:53:06Z)
• Gated Multi-Resolution Transfer Network for Burst Restoration and Enhancement [75.25451566988565]
  We propose a novel Gated Multi-Resolution Transfer Network (GMTNet) to reconstruct a spatially precise high-quality image from a burst of low-quality raw images. Detailed experimental analysis on five datasets validates our approach and sets a state-of-the-art for burst super-resolution, burst denoising, and low-light burst enhancement.
  arXiv  Detail & Related papers  (2023-04-13T17:54:00Z)
• DDT: Dual-branch Deformable Transformer for Image Denoising [6.596462333804802]
  Transformer is beneficial for image denoising tasks since it can model long-range dependencies to overcome limitations presented by inductive convolutional biases. We propose an efficient Dual-branch Deformable Transformer (DDT) denoising network which captures both local and global interactions in parallel.
  arXiv  Detail & Related papers  (2023-04-13T08:54:44Z)
• DCS-RISR: Dynamic Channel Splitting for Efficient Real-world Image Super-Resolution [15.694407977871341]
  Real-world image super-resolution (RISR) has received increased focus for improving the quality of SR images under unknown complex degradation. Existing methods rely on the heavy SR models to enhance low-resolution (LR) images of different degradation levels. We propose a novel Dynamic Channel Splitting scheme for efficient Real-world Image Super-Resolution, termed DCS-RISR.
  arXiv  Detail & Related papers  (2022-12-15T04:34:57Z)
• Sub-Aperture Feature Adaptation in Single Image Super-resolution Model for Light Field Imaging [17.721259583120396]
  This paper proposes an adaptation module in a pretrained Single Image Super Resolution (SISR) network to leverage the powerful SISR model. It is an adaptation in the SISR network to further exploit the spatial and angular information in LF images to improve the super resolution performance.
  arXiv  Detail & Related papers  (2022-07-25T03:43:56Z)
• FCL-GAN: A Lightweight and Real-Time Baseline for Unsupervised Blind Image Deblurring [72.43250555622254]
  We propose a lightweight and real-time unsupervised BID baseline, termed Frequency-domain Contrastive Loss Constrained Lightweight CycleGAN. FCL-GAN has attractive properties, i.e., no image domain limitation, no image resolution limitation, 25x lighter than SOTA, and 5x faster than SOTA. Experiments on several image datasets demonstrate the effectiveness of FCL-GAN in terms of performance, model size and reference time.
  arXiv  Detail & Related papers  (2022-04-16T15:08:03Z)
• Efficient and Degradation-Adaptive Network for Real-World Image Super-Resolution [28.00231586840797]
  Real-world image super-resolution (Real-ISR) is a challenging task due to the unknown complex degradation of real-world images. Recent research on Real-ISR has achieved significant progress by modeling the image degradation space. We propose an efficient degradation-adaptive super-resolution (DASR) network, whose parameters are adaptively specified by estimating the degradation of each input image.
  arXiv  Detail & Related papers  (2022-03-27T05:59:13Z)
• Rich CNN-Transformer Feature Aggregation Networks for Super-Resolution [50.10987776141901]
  Recent vision transformers along with self-attention have achieved promising results on various computer vision tasks. We introduce an effective hybrid architecture for super-resolution (SR) tasks, which leverages local features from CNNs and long-range dependencies captured by transformers. Our proposed method achieves state-of-the-art SR results on numerous benchmark datasets.
  arXiv  Detail & Related papers  (2022-03-15T06:52:25Z)
• Global Vision Transformer Pruning with Hessian-Aware Saliency [93.33895899995224]
  This work challenges the common design philosophy of the Vision Transformer (ViT) model with uniform dimension across all the stacked blocks in a model stage. We derive a novel Hessian-based structural pruning criteria comparable across all layers and structures, with latency-aware regularization for direct latency reduction. Performing iterative pruning on the DeiT-Base model leads to a new architecture family called NViT (Novel ViT), with a novel parameter that utilizes parameters more efficiently.
  arXiv  Detail & Related papers  (2021-10-10T18:04:59Z)
• Deep Burst Super-Resolution [165.90445859851448]
  We propose a novel architecture for the burst super-resolution task. Our network takes multiple noisy RAW images as input, and generates a denoised, super-resolved RGB image as output. In order to enable training and evaluation on real-world data, we additionally introduce the BurstSR dataset.
  arXiv  Detail & Related papers  (2021-01-26T18:57:21Z)
• Boosting Image Super-Resolution Via Fusion of Complementary Information Captured by Multi-Modal Sensors [21.264746234523678]
  Image Super-Resolution (SR) provides a promising technique to enhance the image quality of low-resolution optical sensors. In this paper, we attempt to leverage complementary information from a low-cost channel (visible/depth) to boost image quality of an expensive channel (thermal) using fewer parameters.
  arXiv  Detail & Related papers  (2020-12-07T02:15:28Z)
• Multi-Dimension Fusion Network for Light Field Spatial Super-Resolution using Dynamic Filters [23.82780431526054]
  We introduce a novel learning-based framework to improve the spatial resolution of light fields. Our reconstructed images also show sharp details and distinct lines in both sub-aperture images and epipolar plane images.
  arXiv  Detail & Related papers  (2020-08-26T09:05:07Z)
• Light Field Spatial Super-resolution via Deep Combinatorial Geometry Embedding and Structural Consistency Regularization [99.96632216070718]
  Light field (LF) images acquired by hand-held devices usually suffer from low spatial resolution. The high-dimensional spatiality characteristic and complex geometrical structure of LF images make the problem more challenging than traditional single-image SR. We propose a novel learning-based LF framework, in which each view of an LF image is first individually super-resolved.
  arXiv  Detail & Related papers  (2020-04-05T14:39:57Z)
• Learning Light Field Angular Super-Resolution via a Geometry-Aware Network [101.59693839475783]
  We propose an end-to-end learning-based approach aiming at angularly super-resolving a sparsely-sampled light field with a large baseline. Our method improves the PSNR of the second best method up to 2 dB in average, while saves the execution time 48$times$.
  arXiv  Detail & Related papers  (2020-02-26T02:36:57Z)

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.