DiffIR: Efficient Diffusion Model for Image Restoration

• URL: http://arxiv.org/abs/2303.09472v3
• Date: Wed, 16 Aug 2023 14:36:41 GMT
• Title: DiffIR: Efficient Diffusion Model for Image Restoration
• Authors: Bin Xia, Yulun Zhang, Shiyin Wang, Yitong Wang, Xinglong Wu, Yapeng Tian, Wenming Yang, and Luc Van Gool
• Abstract summary: Diffusion model (DM) has achieved SOTA performance by modeling the image synthesis process into a sequential application of a denoising network. Traditional DMs running massive iterations on a large model to estimate whole images or feature maps is inefficient for image restoration. We propose DiffIR, which consists of a compact IR prior extraction network (CPEN), dynamic IR transformer (DIRformer), and denoising network.
• Score: 108.82579440308267
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Diffusion model (DM) has achieved SOTA performance by modeling the image synthesis process into a sequential application of a denoising network. However, different from image synthesis, image restoration (IR) has a strong constraint to generate results in accordance with ground-truth. Thus, for IR, traditional DMs running massive iterations on a large model to estimate whole images or feature maps is inefficient. To address this issue, we propose an efficient DM for IR (DiffIR), which consists of a compact IR prior extraction network (CPEN), dynamic IR transformer (DIRformer), and denoising network. Specifically, DiffIR has two training stages: pretraining and training DM. In pretraining, we input ground-truth images into CPEN$_{S1}$ to capture a compact IR prior representation (IPR) to guide DIRformer. In the second stage, we train the DM to directly estimate the same IRP as pretrained CPEN$_{S1}$ only using LQ images. We observe that since the IPR is only a compact vector, DiffIR can use fewer iterations than traditional DM to obtain accurate estimations and generate more stable and realistic results. Since the iterations are few, our DiffIR can adopt a joint optimization of CPEN$_{S2}$, DIRformer, and denoising network, which can further reduce the estimation error influence. We conduct extensive experiments on several IR tasks and achieve SOTA performance while consuming less computational costs. Code is available at \url{https://github.com/Zj-BinXia/DiffIR}.

Related papers

• Invertible Residual Rescaling Models [46.28263683643467]
  Invertible Rescaling Networks (IRNs) and their variants have witnessed remarkable achievements in various image processing tasks like image rescaling. We propose Invertible Residual Rescaling Models (IRRM) for image rescaling by learning a bijection between a high-resolution image and its low-resolution counterpart with a specific distribution. Our IRRM has respectively PSNR gains of at least 0.3 dB over HCFlow and IRN in the x4 rescaling while only using 60% parameters and 50% FLOPs.
  arXiv  Detail & Related papers  (2024-05-05T14:14:49Z)
• Modality Translation for Object Detection Adaptation Without Forgetting Prior Knowledge [11.905387325966311]
  This paper focuses on adapting a large object detection model trained on RGB images to new data extracted from IR images. We propose Modality Translator (ModTr) as an alternative to the common approach of fine-tuning a large model to the new modality.
  arXiv  Detail & Related papers  (2024-04-01T21:28:50Z)
• Learning to See Low-Light Images via Feature Domain Adaptation [17.033219611079165]
  We propose a single-stage network empowered by Feature Domain Adaptation (FDA) to decouple the denoising and color mapping tasks in raw LLIE. FDA can explore the global and local correlations with fewer line buffers. Our method achieves state-of-the-art performance with fewer computing costs.
  arXiv  Detail & Related papers  (2023-12-11T03:38:26Z)
• EDiffSR: An Efficient Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution [32.956539422513416]
  convolutional networks have achieved remarkable development in remote sensing image Super-Resoltuion (SR) Generative adversarial networks have the potential to infer intricate details, but they are easy to collapse, resulting in undesirable artifacts. EDiffSR is easy to train and maintains the merits of DPM in generating perceptual-pleasant images.
  arXiv  Detail & Related papers  (2023-10-30T06:09:33Z)
• DiffI2I: Efficient Diffusion Model for Image-to-Image Translation [108.82579440308267]
  Diffusion Model (DM) has emerged as the SOTA approach for image synthesis. DM can't perform well on some image-to-image translation (I2I) tasks. DiffI2I comprises three key components: a compact I2I prior extraction network (CPEN), a dynamic I2I transformer (DI2Iformer) and a denoising network.
  arXiv  Detail & Related papers  (2023-08-26T05:18:23Z)
• Hierarchical Integration Diffusion Model for Realistic Image Deblurring [71.76410266003917]
  Diffusion models (DMs) have been introduced in image deblurring and exhibited promising performance. We propose the Hierarchical Integration Diffusion Model (HI-Diff), for realistic image deblurring. Experiments on synthetic and real-world blur datasets demonstrate that our HI-Diff outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-05-22T12:18:20Z)
• Denoising Diffusion Models for Plug-and-Play Image Restoration [135.6359475784627]
  This paper proposes DiffPIR, which integrates the traditional plug-and-play method into the diffusion sampling framework. Compared to plug-and-play IR methods that rely on discriminative Gaussian denoisers, DiffPIR is expected to inherit the generative ability of diffusion models.
  arXiv  Detail & Related papers  (2023-05-15T20:24:38Z)
• Towards Lightweight Super-Resolution with Dual Regression Learning [58.98801753555746]
  Deep neural networks have exhibited remarkable performance in image super-resolution (SR) tasks. The SR problem is typically an ill-posed problem and existing methods would come with several limitations. We propose a dual regression learning scheme to reduce the space of possible SR mappings.
  arXiv  Detail & Related papers  (2022-07-16T12:46:10Z)

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.