《Diffusion Models Paper》1.0 经典必读论文10篇

        最近开启了论文100计划，准备阅读整理100篇CV领域经典论文，在扩散模型方面精选了十篇经典论文，这里相当于一个目录，后续会对这十篇论文一一详细精度

目录

一.《Improved Denoising Diffusion Probabilistic Models》DDPM奠基之作

二.《DENOISING DIFFUSION IMPLICIT MODELS》从DDPM到DDIM

三.《Diffusion Models Beat GANs on Image Classification》Diffusion第一次击败GANs

四.《Classifier-Free Diffusion Guidance》条件分类器技术

五.《Palette: Image-to-Image Diffusion Models》Image-to-Image经典之作

六.《GLIDE》畅游多模态领域

七.《High-Resolution Image Synthesis with Latent Diffusion Models》Stable Diffusion

八.《Video Diffusion Models》进军视频领域

九.《Prompt-to-Prompt Image Editing with Cross Attention Control》自注意机制

十.《Scalable Diffusion Models with Transformers》Transformer instead Unet

---

一.《Improved Denoising Diffusion Probabilistic Models》DDPM奠基之作

1.博客:

2.论文:http://proceedings.mlr.press/v139/nichol21a/nichol21a.pdf

3.代码:https://github.com/lucidrains/denoising-diffusion-pytorch

                  https://github.com/openai/improved-diffusion

二.《DENOISING DIFFUSION IMPLICIT MODELS》从DDPM到DDIM

1.博客:

2.论文:https://arxiv.org/pdf/2010.02502.pdf

3.代码:https://github.com/ermongroup/ddim

三.《Diffusion Models Beat GANs on Image Classification》Diffusion第一次击败GANs

1.博客:

2.论文:https://arxiv.org/pdf/2307.08702.pdf

3.代码:GitHub - openai/guided-diffusion

四.《Classifier-Free Diffusion Guidance》条件分类器技术

1.博客:

2.论文:https://openreview.net/pdf?id=qw8AKxfYbI

3.代码:GitHub - lucidrains/classifier-free-guidance-pytorch: Implementation of Classifier Free Guidance in Pytorch, with emphasis on text conditioning, and flexibility to include multiple text embedding models

五.《Palette: Image-to-Image Diffusion Models》Image-to-Image经典之作

1.博客:

2.论文:https://dl.acm.org/doi/pdf/10.1145/3528233.3530757

3.代码:GitHub - Janspiry/Palette-Image-to-Image-Diffusion-Models: Unofficial implementation of Palette: Image-to-Image Diffusion Models by Pytorch

六.《GLIDE》畅游多模态领域

1.博客:

2.论文:https://arxiv.org/pdf/2112.10741.pdf

3.代码:https://github.com/openai/glide-text2im/tree/main/glide_text2im

七.《High-Resolution Image Synthesis with Latent Diffusion Models》Stable Diffusion

1.博客:

2.论文:https://arxiv.org/pdf/2112.10752.pdf

3.代码:https://arxiv.org/pdf/2112.10752.pdf

八.《Video Diffusion Models》进军视频领域

1.博客:

2.论文:https://arxiv.org/pdf/2204.03458.pdf

3.代码:GitHub - lucidrains/video-diffusion-pytorch: Implementation of Video Diffusion Models, Jonathan Ho's new paper extending DDPMs to Video Generation - in Pytorch

九.《Prompt-to-Prompt Image Editing with Cross Attention Control》自注意机制

1.博客:

2.论文:https://arxiv.org/pdf/2208.01626.pdf

3.代码:https://github.com/bloc97/CrossAttentionControl

十.《Scalable Diffusion Models with Transformers》Transformer instead Unet

1.博客:

2.论文:https://arxiv.org/pdf/2212.09748.pdf

3.代码:https://github.com/facebookresearch/DiT