Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[ICCV 2023] "TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition" (Official Implementation)
| Date | Stars |
|---|---|
| 2026-07-24 | 814 |
| 2026-07-25 | 814 |
| 2026-07-28 | 814 |
| 2026-07-30 | 814 |
| 2026-08-06 | 814 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition (ICCV 2023)
## [<a href="https://shilin-lu.github.io/tf-icon.github.io/" target="_blank">Project Page</a>] [<a href="https://entuedu-my.sharepoint.com/:b:/g/personal/shilin002_e_ntu_edu_sg/EWRDLuFDrs5Ll0KGuMtvtbUBhBZcSw2roKCo96iCWgpMZQ?e=rEv3As" target="_blank">Poster</a>]
[](https://arxiv.org/abs/2307.12493) [](https://entuedu-my.sharepoint.com/:f:/g/personal/shilin002_e_ntu_edu_sg/EmmCgLm_3OZCssqjaGdvjMwBCIvqfjsyphjqNs7g2DFzQQ?e=JSwOHY)
Official implementation of [TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition](https://shilin-lu.github.io/tf-icon.github.io/).
> **TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition**<br>
<!-- > [Gwanghyun Kim](https://gwang-kim.github.io/), Taesung Kwon, [Jong Chul Ye](https://bispl.weebly.com/professor.html) <br> -->
> Shilin Lu, Yanzhu Liu, and Adams Wai-Kin Kong <br>
> ICCV 2023
>
>**Abstract**: <br>
Text-driven diffusion models have exhibited impressive generative capabilities, enabling various image editing tasks. In this paper, we propose TF-ICON, a novel Training-Free Image COmpositioN framework that harnesses the power of text-driven diffusion models for cross-domain image-guided composition. This task aims to seamlessly integrate user-provided objects into a specific visual context. Current diffusion-based methods often involve costly instance-based optimization or finetuning of pretrained models on customized datasets, which can potentially undermine their rich prior. In contrast, TF-ICON can leverage off-the-shelf diffusion models to perform cross-domain image-guided composition without requiring additional training, finetuning, or optimization. Moreover, we introduce the exceptional prompt, which contains no information, to facilitate text-driven diffusion models in accurately inverting real images into latent representations, forming the basis for compositing. Our experiments show that equipping Stable Diffusion with the exceptional prompt outperforms state-of-the-art inversion methods on various datasets (CelebA-HQ, COCO, and ImageNet), and that TF-ICON surpasses prior baselines in versatile visual domains.
<!-- ## [<a href="https://pnp-diffusion.github.io/" target="_blank">Project Page</a>] [<a href="https://github.com/MichalGeyer/pnp-diffusers" target="_blank">Diffusers Implementation</a>] -->
<!-- [](https://arxiv.org/abs/2211.12572) [](https://huggingface.co/spaces/hysts/PnP-diffusion-features) <a href="https://replicate.com/arielreplicate/plug_and_play_image_translation"><img src="https://replicate.com/arielreplicate/plug_and_play_image_translation/badge"></a> [](https://www.dropbox.com/sh/8giw0uhfekft47h/AAAF1frwakVsQocKczZZSX6La?dl=0) -->
---
</div>
<!-- # Updates:
**19/06/23** 🧨 Diffusers implementation of Plug-and-Play is available [here](https://github.com/MichalGeyer/pnp-diffusers). -->
<!-- ## TODO:
- [ ] Diffusers support and pipeline integration
- [ ] Gradio demo
- [ ] Release TF-ICON Test Benchmark -->
<!-- ## Usage
**To plug-and-play diffusion features, please follow these steps:**
1. [Setup](#setup)
2. [Feature extraction](#feature-extraction)
3. [Running PnP](#running-pnp)
4. [TI2I Benchmarks](#ti2i-benchmarks) -->
---
</div>
## Contents
- [Setup](#setup)
- [Option 1: Using Conda](#option-1-using-conda)
- [Option 2: Using Pip with Virtual Environment](#option-2-using-pip-with-virtual-environment)
- [Option 3: Using Pip (Global Installation)](#option-3-using-pip-global-installation)
Excerpt of 11,443 characters
Read on GitHub
48
2
Ikko Eltociear Ashimine · Japan
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:363b3cb19a121f8e, topic:stable-diffusion, topic:text-to-image, readme:stable diffusion