Inf-DiT: Upsampling Any-Resolution Image with Memory-Efficient Diffusion Transformer

• URL: http://arxiv.org/abs/2405.04312v2
• Date: Wed, 8 May 2024 07:17:12 GMT
• Title: Inf-DiT: Upsampling Any-Resolution Image with Memory-Efficient Diffusion Transformer
• Authors: Zhuoyi Yang, Heyang Jiang, Wenyi Hong, Jiayan Teng, Wendi Zheng, Yuxiao Dong, Ming Ding, Jie Tang,
• Abstract summary: In this work, we propose a unidirectional block attention mechanism that can adjust the memory overhead during the inference process. Building on this module, we adopt the DiT structure for upsampling and develop an infinite super-resolution model. Our model achieves SOTA performance in generating ultra-high-resolution images in both machine and human evaluation.
• Score: 36.715376148053785
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion models have shown remarkable performance in image generation in recent years. However, due to a quadratic increase in memory during generating ultra-high-resolution images (e.g. 4096*4096), the resolution of generated images is often limited to 1024*1024. In this work. we propose a unidirectional block attention mechanism that can adaptively adjust the memory overhead during the inference process and handle global dependencies. Building on this module, we adopt the DiT structure for upsampling and develop an infinite super-resolution model capable of upsampling images of various shapes and resolutions. Comprehensive experiments show that our model achieves SOTA performance in generating ultra-high-resolution images in both machine and human evaluation. Compared to commonly used UNet structures, our model can save more than 5x memory when generating 4096*4096 images. The project URL is https://github.com/THUDM/Inf-DiT.

Related papers

• ZoomLDM: Latent Diffusion Model for multi-scale image generation [57.639937071834986]
  We present ZoomLDM, a diffusion model tailored for generating images across multiple scales. Central to our approach is a novel magnification-aware conditioning mechanism that utilizes self-supervised learning (SSL) embeddings. ZoomLDM achieves state-of-the-art image generation quality across all scales, excelling in the data-scarce setting of generating thumbnails of entire large images.
  arXiv  Detail & Related papers  (2024-11-25T22:39:22Z)
• Alleviating Distortion in Image Generation via Multi-Resolution Diffusion Models [26.926712014346432]
  This paper presents innovative enhancements to diffusion models by integrating a novel multi-resolution network and time-dependent layer normalization. Our method's efficacy is demonstrated on the class-conditional ImageNet generation benchmark, setting new state-of-the-art FID scores of 1.70 on ImageNet 256 x 256 and 2.89 on ImageNet 512 x 512.
  arXiv  Detail & Related papers  (2024-06-13T17:59:58Z)
• HiDiffusion: Unlocking Higher-Resolution Creativity and Efficiency in Pretrained Diffusion Models [13.68666823175341]
  HiDiffusion is a tuning-free higher-resolution framework for image synthesis. RAU-Net dynamically adjusts the feature map size to resolve object duplication. MSW-MSA engages optimized window attention to reduce computations.
  arXiv  Detail & Related papers  (2023-11-29T11:01:38Z)
• Matryoshka Diffusion Models [38.26966802461602]
  Diffusion models are the de facto approach for generating high-quality images and videos. We introduce Matryoshka Diffusion Models, an end-to-end framework for high-resolution image and video synthesis. We demonstrate the effectiveness of our approach on various benchmarks, including class-conditioned image generation, high-resolution text-to-image, and text-to-video applications.
  arXiv  Detail & Related papers  (2023-10-23T17:20:01Z)
• ScaleCrafter: Tuning-free Higher-Resolution Visual Generation with Diffusion Models [126.35334860896373]
  We investigate the capability of generating images from pre-trained diffusion models at much higher resolutions than the training image sizes. Existing works for higher-resolution generation, such as attention-based and joint-diffusion approaches, cannot well address these issues. We propose a simple yet effective re-dilation that can dynamically adjust the convolutional perception field during inference.
  arXiv  Detail & Related papers  (2023-10-11T17:52:39Z)
• Any-Size-Diffusion: Toward Efficient Text-Driven Synthesis for Any-Size HD Images [56.17404812357676]
  Stable diffusion, a generative model used in text-to-image synthesis, frequently encounters composition problems when generating images of varying sizes. We propose a two-stage pipeline named Any-Size-Diffusion (ASD), designed to efficiently generate well-composed images of any size. We show that ASD can produce well-structured images of arbitrary sizes, cutting down the inference time by 2x compared to the traditional tiled algorithm.
  arXiv  Detail & Related papers  (2023-08-31T09:27:56Z)
• CUF: Continuous Upsampling Filters [25.584630142930123]
  In this paper, we consider one of the most important operations in image processing: upsampling. We propose to parameterize upsampling kernels as neural fields. This parameterization leads to a compact architecture that obtains a 40-fold reduction in the number of parameters when compared with competing arbitrary-scale super-resolution architectures.
  arXiv  Detail & Related papers  (2022-10-13T12:45:51Z)
• Any-resolution Training for High-resolution Image Synthesis [55.19874755679901]
  Generative models operate at fixed resolution, even though natural images come in a variety of sizes. We argue that every pixel matters and create datasets with variable-size images, collected at their native resolutions. We introduce continuous-scale training, a process that samples patches at random scales to train a new generator with variable output resolutions.
  arXiv  Detail & Related papers  (2022-04-14T17:59:31Z)
• Cascaded Diffusion Models for High Fidelity Image Generation [53.57766722279425]
  We show that cascaded diffusion models are capable of generating high fidelity images on the class-conditional ImageNet generation challenge. A cascaded diffusion model comprises a pipeline of multiple diffusion models that generate images of increasing resolution. We find that the sample quality of a cascading pipeline relies crucially on conditioning augmentation.
  arXiv  Detail & Related papers  (2021-05-30T17:14:52Z)
• Memory-efficient GAN-based Domain Translation of High Resolution 3D Medical Images [0.15092198588928965]
  Generative adversarial networks (GANs) are rarely applied on 3D medical images of large size. The present work proposes a multi-scale patch-based GAN approach for establishing unpaired domain translation. The evaluation of the domain translation scenarios is performed on brain MRIs of size 155x240x240 and thorax CTs of size up to 512x512x512.
  arXiv  Detail & Related papers  (2020-10-06T08:43:27Z)

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.