diff-usion / awesome-diffusion-models Goto Github PK
View Code? Open in Web Editor NEWA collection of resources and papers on Diffusion Models
Home Page: https://diff-usion.github.io/Awesome-Diffusion-Models/
License: MIT License
A collection of resources and papers on Diffusion Models
Home Page: https://diff-usion.github.io/Awesome-Diffusion-Models/
License: MIT License
Hi heejkoo,
The paper "DisDiff: Unsupervised Disentanglement of Diffusion Probabilistic Models" should belong to Vision/Generation but not 3D vision instead, because the paper is focus on learning disentangled representation of 2D images.
Thanks,
Thomas
Hi,
Please consider including our recent work, SelfRDB, in your collection. Here are the links to the preprint and the GitHub repository:
ArXiv preprint: https://arxiv.org/abs/2405.06789
GitHub repo: https://github.com/icon-lab/SelfRDB
Thank you for this collection.
Thank you for constructing such a great repo! We have a update for paper listed: Could you do some modification to the paper "A Flexible Diffusion Model" by Weitao Du. It has been accepted by ICML 2023.
I suggest you add a new survey paper: https://arxiv.org/abs/2303.07909. Thank you!
I’m delighted that our paper ‘DocDiff: Document Enhancement via Residual Diffusion Models’ has been included in this repository. However, there seems to be an issue with its classification. It should not be placed under the Natural Language category, but rather under Inverse Problems within the Vision category.
Hello! Thank you for your hard work on this Repo.
We are the authors of this diffusion-based video editing paper with bounding box. It would be great to be listed here.
[Paper] TrailBlazer: Trajectory Control for Diffusion-Based Video Generation
[Project link] https://hohonu-vicml.github.io/Trailblazer.Page/
introductory video with coding example
I have recently published two research papers on time series generation models on arXiv. I came across your excellent repository, Awesome-Diffusion-Models, on GitHub and have found it to be an incredibly valuable resource for researchers in the field.
I would like to bring to your attention two papers that I have uploaded, which I believe would be a great fit for your collection. The content of these papers aligns well with the themes of your project and may be of interest to other contributors and users.
Would it be possible to include links to these papers in your repository? Here are the arXiv links for easy reference:
Diff-MTS: Temporal-Augmented Conditional Diffusion-based AIGC for Industrial Time Series Towards the Large Model Era, arXiv:2407.11501.
AIGC for Industrial Time Series: From Deep Generative Models to Large Generative Models,arXiv:2407.11480.
Hi,
Thanks for your wonderful survey!
Would you mind adding our paper published in EMNLP 2023 in industry paper section? It leverages diffusion for topic guided text generation.
Title: DeTiME: Diffusion-Enhanced Topic Modeling using Encoder-decoder based LLM
Paper link: https://arxiv.org/abs/2310.15296
The Schrodinger bridge problem is a question from statistical physics to find the shortest path that evolves a dynamic system into another system. (I'm not familiar with the concepts, though)
https://arxiv.org/abs/1608.05862
And here are some papers that relate this problem with the diffusion probabilistic model.
https://arxiv.org/abs/2106.01357
http://proceedings.mlr.press/v139/wang21l/wang21l.pdf
Thank you so much for this excellent repo you built!
Could you please add the link https://heejkoo.github.io/Awesome-Diffusion-Models/ to the About description in this repo? It will be very helpful.
Hi. I noticed that the paper "DiffWave: a versatile diffusion model for speech synthesis" has wrong author list and conference information. Also, there is no paper titled "DiffWave with Continuous-time Variational Diffusion Models" to my knowledge. Could you fix that? Thanks!
Could you do some modification to the paper "Diffusion Motion: Generate Text-Guided 3D Human Motion by Diffusion Model" by Zhiyuan Ren. It has been accepted by ICASSP 2023 yesterday.
Our paper on diffusion Schrodinger bridges was one of the first to my knowledge doing image to image translation with diffusion models
https://arxiv.org/abs/2106.01357.
Not sure how categories are defined here.
Under Segmentation, the following paper is listed:
Domain-agnostic segmentation of thalamic nuclei from joint structural and diffusion MRI
Henry F. J. Tregidgo et al. 2023
This work does not actually use a diffusion model, but a straightforward 3D U-Net to segment brain structures.
The confusion is probably caused by the concept of Diffusion MRI. This sequence, simply put, tracks the direction and velocity of fluids (mostly water) moving around a given spatial location. Most often this is done in the brain as it's quite easy to keep in place with little to no patient movement. This lets you create these cool looking fiber tractography images, as fluid is more likely to move in parallel to the nerve bundles spread across your brain!
Hello, thanks for your great repo, it is really awesome. Recently we published our new paper "Diffusion-Based Scene Graph to Image Generation with Masked Contrastive Pre-Training" on arXiv, which proposes a diffusion model that directly conditions on scene graphs to generate images. Could you please add this paper to the repo? Thanks a lot!
arxiv: https://arxiv.org/abs/2211.11138
GitHub: https://github.com/YangLing0818/SGDiff
Back to Optimization: Diffusion-based Zero-Shot 3D Human Pose Estimation
https://arxiv.org/abs/2307.03833
Hi @heejkoo , i think we can include diffusion models from the image forensic perspective, which states how to distinguish image generated by diffusion model from real images. This is a meaningful research direction and has many practical needs in the security system, as well as helps people use diffusion model to generate more "real" images. This image forensic has many topics, such as localization, detection, and attribution. Please consider taking a look on these following works:
Towards the Detection of Diffusion Model Deepfakes (https://arxiv.org/pdf/2210.14571.pdf)
Hierarchical Fine-Grained Image Forgery Detection and Localization (CVPR2023) (https://arxiv.org/pdf/2303.17111.pdf)
AutoSplice: A Text-prompt Manipulated Image Dataset for Media Forensics (https://arxiv.org/pdf/2304.06870.pdf)
DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models (https://arxiv.org/pdf/2210.06998.pdf)
Hi, @heejkoo,
Thanks very much for your efforts in collecting these papers! I am wondering whether you can add our latest cvpr paper about the image-to-video generation with diffusion models.
Title: Conditional Image-to-Video Generation with Latent Flow Diffusion Models.
Paper Link: https://arxiv.org/pdf/2303.13744.pdf
Code Link: https://github.com/nihaomiao/CVPR23_LFDM
Thanks a lot!
Hi,
We have a new paper on graph generation. Can you please add it to the README?
**SwinGNN: Rethinking Permutation Invariance in Diffusion Models for Graph Generation** \
*Qi Yan, Zhengyang Liang, Yang Song, Renjie Liao, Lele Wang* \
arXiv 2023. [[Paper](https://arxiv.org/abs/2307.01646)] [[Github](https://github.com/qiyan98/SwinGNN)] \
4 Jul 2023
It's also in the pull request.
Thanks.
Thank you for such a great repo!
Would it be possible for you to add the following ICML 2022 paper on graph generation (and also related to molecule generation)?
Score-based Generative Modeling of Graphs via the System of Stochastic Differential Equations, Jo et al., ICML 2022
paper: https://arxiv.org/abs/2202.02514
github: https://github.com/harryjo97/GDSS
The name of the paper "First Hitting Diffusion Models" is "First Hitting Diffusion Models for Generating Manifold, Graph and Categorical Data" actually.
Thank you for constructing such a great repo! We have a update for paper listed: Could you do some modification to the paper "DisDiff: Unsupervised Disentanglement of Diffusion Probabilistic Models" by Tao Yang. It has been accepted by NeurIPS 2023.
Hi @heejkoo,
"Semantic Diffusion Network for Semantic Segmentation" is not technically a Diffusion based generative model paper. I think the title is a little misleading in some sense.
If possible, you can check and can remove from the list.
Thanks!
Hi, thanks for providing such a meaningful survey project. Could you please help to add the following paper:
Paper name: VideoComposer: Compositional Video Synthesis with Motion Controllability (NeurIPS-2023)
Paper link: https://arxiv.org/abs/2306.02018
Project link: https://videocomposer.github.io/
Code link: https://github.com/damo-vilab/videocomposer
Related area: Video Generation, Controllable Video Generation
Thank you!
Paper name: Generative Prompt Model for Weakly Supervised Object Localization
Paper link: https://arxiv.org/abs/2307.09756
Code link: https://github.com/callsys/GenPromp
Related area: Segmentation
Thank you!
Deep diffusion-based forecasting of COVID-19 by incorporating network-level mobility information
This paper seems not related to diffusion model, where "diffusion" means the spreading process of pandemic in this paper
Hey,
Thanks for your amazing work keeping this repository updated! I wanted to ask you to add this ICCV'23 paper in the 3D Vision section. It was there at some point but disappeared :)
BeLFusion: Latent Diffusion for Behavior-Driven Human Motion Prediction
Project page: https://barquerogerman.github.io/BeLFusion/
Thanks!!
Hi! Could you please include our work StableVideo at ICCV 2023? Thank you so much!
StableVideo: Text-driven Consistency-aware Diffusion Video Editing
Wenhao Chai, Xun Guo, Gaoang Wang, Yan Lu
ICCV 2023
https://arxiv.org/abs/2308.09592
Hi, thanks for making this great repo to track the recent awesome works.
We have recently posted our paper to arXiv (https://arxiv.org/abs/2205.12952) with the webpage (https://tengfei-wang.github.io/PITI/index.html), which uses diffusion model for image-to-image translation. Could you please include this paper in the repo?
Thanks!
Maximum diffusion reinforcement learning
https://arxiv.org/pdf/2309.15293
Can you please add our diffusion paper on medical image reconstruction to your repository.
Learning Fourier-Constrained Diffusion Bridges for MRI Reconstruction
Hi! Just wanted to share this new audio generation paper, which I believe is relevant. Thanks!
CVPR 2022 Tutorial - Denoising Diffusion-based Generative Modeling: Foundations and Applications
Video link: https://www.youtube.com/watch?v=cS6JQpEY9cs
Hi -- our paper "Realistic galaxy image simulation via score-based generative models" applies a diffusion model to galaxy simulation.
Here's the arxiv link:
https://arxiv.org/abs/2111.01713
You may consider adding this paper: Composable Text Controls in Latent Space with ODEs
https://arxiv.org/abs/2208.00638
And the code is released at https://github.com/guangyliu/LatentOps
Hi, thank you for your great repo.
We have recently published our survey paper, "Diffusion Models for Medical Image Analysis: A Comprehensive Survey," to arXiv. Could you please add this paper to the repo? Thank you very much.
Arxiv: https://arxiv.org/abs/2211.07804
GitHub: https://github.com/amirhossein-kz/Awesome-Diffusion-Models-in-Medical-Imaging
Hi, @heejkoo,
Thanks very much for your efforts in collecting these papers! I was wondering could you please add our recent diffusion paper for 3D Brain MRI generartion using 2D diffusion model?
Title: Generating Realistic 3D Brain MRIs Using a Conditional Diffusion Probabilistic Model
Authors: Wei Peng, Ehsan Adeli, Qingyu Zhao, Kilian M Pohl
Paper Link: https://arxiv.org/pdf/2212.08034
Code Link: https://github.com/Project-MONAI/GenerativeModels/tree/260-add-cdpm-model
Thanks a lot!
Paper Title: Effective Real Image Editing with Accelerated Iterative Diffusion Inversion
Paper link: https://arxiv.org/abs/2309.04907
Related area: Image Editing, Image Generation
Thanks
Please add our paper on SDE-based vocoder. Thank you.
The paper https://arxiv.org/abs/2201.12519; the demo: https://wushoule.github.io/ItoAudio/.
The first paper of diffusion models in instance segmentation:
DiffusionInst: Diffusion Model for Instance Segmentation
arxiv: https://arxiv.org/abs/2212.02773
code: https://github.com/chenhaoxing/DiffusionInst
Nice job!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.