LoFormer: Local Frequency Transformer for Image Deblurring

• URL: http://arxiv.org/abs/2407.16993v1
• Date: Wed, 24 Jul 2024 04:27:03 GMT
• Title: LoFormer: Local Frequency Transformer for Image Deblurring
• Authors: Xintian Mao, Jiansheng Wang, Xingran Xie, Qingli Li, Yan Wang,
• Abstract summary: We introduce a novel approach termed Local Frequency Transformer (LoFormer) Within each unit of LoFormer, we incorporate a Local Channel-wise SA in the frequency domain (Freq-LC) to simultaneously capture cross-covariance within low- and high-frequency local windows. Our experiments demonstrate that LoFormer significantly improves performance in the image deblurring task, achieving a PSNR of 34.09 dB on the GoPro dataset with 126G FLOPs.
• Score: 12.032239441930306
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Due to the computational complexity of self-attention (SA), prevalent techniques for image deblurring often resort to either adopting localized SA or employing coarse-grained global SA methods, both of which exhibit drawbacks such as compromising global modeling or lacking fine-grained correlation. In order to address this issue by effectively modeling long-range dependencies without sacrificing fine-grained details, we introduce a novel approach termed Local Frequency Transformer (LoFormer). Within each unit of LoFormer, we incorporate a Local Channel-wise SA in the frequency domain (Freq-LC) to simultaneously capture cross-covariance within low- and high-frequency local windows. These operations offer the advantage of (1) ensuring equitable learning opportunities for both coarse-grained structures and fine-grained details, and (2) exploring a broader range of representational properties compared to coarse-grained global SA methods. Additionally, we introduce an MLP Gating mechanism complementary to Freq-LC, which serves to filter out irrelevant features while enhancing global learning capabilities. Our experiments demonstrate that LoFormer significantly improves performance in the image deblurring task, achieving a PSNR of 34.09 dB on the GoPro dataset with 126G FLOPs. https://github.com/DeepMed-Lab-ECNU/Single-Image-Deblur

Related papers

• LaMamba-Diff: Linear-Time High-Fidelity Diffusion Models Based on Local Attention and Mamba [54.85262314960038]
  Local Attentional Mamba blocks capture both global contexts and local details with linear complexity. Our model exhibits exceptional scalability and surpasses the performance of DiT across various model scales on ImageNet at 256x256 resolution. Compared to state-of-the-art diffusion models on ImageNet 256x256 and 512x512, our largest model presents notable advantages, such as a reduction of up to 62% GFLOPs.
  arXiv  Detail & Related papers  (2024-08-05T16:39:39Z)
• Empowering Snapshot Compressive Imaging: Spatial-Spectral State Space Model with Across-Scanning and Local Enhancement [51.557804095896174]
  We introduce a State Space Model with Across-Scanning and Local Enhancement, named ASLE-SSM, that employs a Spatial-Spectral SSM for global-local balanced context encoding and cross-channel interaction promoting. Experimental results illustrate ASLE-SSM's superiority over existing state-of-the-art methods, with an inference speed 2.4 times faster than Transformer-based MST and saving 0.12 (M) of parameters.
  arXiv  Detail & Related papers  (2024-08-01T15:14:10Z)
• Frequency-Assisted Mamba for Remote Sensing Image Super-Resolution [49.902047563260496]
  We develop the first attempt to integrate the Vision State Space Model (Mamba) for remote sensing image (RSI) super-resolution. To achieve better SR reconstruction, building upon Mamba, we devise a Frequency-assisted Mamba framework, dubbed FMSR. Our FMSR features a multi-level fusion architecture equipped with the Frequency Selection Module (FSM), Vision State Space Module (VSSM), and Hybrid Gate Module (HGM)
  arXiv  Detail & Related papers  (2024-05-08T11:09:24Z)
• Learning Enriched Features via Selective State Spaces Model for Efficient Image Deblurring [0.0]
  Image deblurring aims to restore a high-quality image from its corresponding blurred. We propose an efficient image deblurring network that leverages selective state spaces model to aggregate enriched and accurate features. Experimental results demonstrate that the proposed method outperforms state-of-the-art approaches on widely used benchmarks.
  arXiv  Detail & Related papers  (2024-03-29T10:40:41Z)
• Optimal Transport Aggregation for Visual Place Recognition [9.192660643226372]
  We introduce SALAD, which reformulates NetVLAD's soft-assignment of local features to clusters as an optimal transport problem. In SALAD, we consider both feature-to-cluster and cluster-to-feature relations and we also introduce a 'dustbin' cluster, designed to selectively discard features deemed non-informative. Our single-stage method surpasses single-stage baselines in public VPR datasets, but also surpasses two-stage methods that add a re-ranking with significantly higher cost.
  arXiv  Detail & Related papers  (2023-11-27T15:46:19Z)
• Recursive Generalization Transformer for Image Super-Resolution [108.67898547357127]
  We propose the Recursive Generalization Transformer (RGT) for image SR, which can capture global spatial information and is suitable for high-resolution images. We combine the RG-SA with local self-attention to enhance the exploitation of the global context. Our RGT outperforms recent state-of-the-art methods quantitatively and qualitatively.
  arXiv  Detail & Related papers  (2023-03-11T10:44:44Z)
• Spatial-Frequency Attention for Image Denoising [22.993509525990998]
  We propose the spatial-frequency attention network (SFANet) to enhance the network's ability in exploiting long-range dependency. Experiments on multiple denoising benchmarks demonstrate the leading performance of SFANet network.
  arXiv  Detail & Related papers  (2023-02-27T09:07:15Z)
• SuperGF: Unifying Local and Global Features for Visual Localization [13.869227429939423]
  SuperGF is a transformer-based aggregation model that operates directly on image-matching-specific local features. We provide implementations of SuperGF using various types of local features, including dense and sparse learning-based or hand-crafted descriptors.
  arXiv  Detail & Related papers  (2022-12-23T13:48:07Z)
• Global Filter Networks for Image Classification [90.81352483076323]
  We present a conceptually simple yet computationally efficient architecture that learns long-term spatial dependencies in the frequency domain with log-linear complexity. Our results demonstrate that GFNet can be a very competitive alternative to transformer-style models and CNNs in efficiency, generalization ability and robustness.
  arXiv  Detail & Related papers  (2021-07-01T17:58:16Z)
• Frequency Consistent Adaptation for Real World Super Resolution [64.91914552787668]
  We propose a novel Frequency Consistent Adaptation (FCA) that ensures the frequency domain consistency when applying Super-Resolution (SR) methods to the real scene. We estimate degradation kernels from unsupervised images and generate the corresponding Low-Resolution (LR) images. Based on the domain-consistent LR-HR pairs, we train easy-implemented Convolutional Neural Network (CNN) SR models.
  arXiv  Detail & Related papers  (2020-12-18T08:25:39Z)

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.