diff --git a/README.md b/README.md index 3b7915a..470fb9e 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,5 @@ # InstructIR ✏️🖼️ -## [High-Quality Image Restoration Following Human Instructions](https://mv-lab.github.io/InstructIR/) +## [High-Quality Image Restoration Following Human Instructions](https://arxiv.org/abs/2401.16468) [![arXiv](https://img.shields.io/badge/arXiv-Paper-.svg)](https://arxiv.org/abs/2401.16468) google colab logo @@ -8,11 +8,15 @@ [![Paper page](https://huggingface.co/datasets/huggingface/badges/resolve/main/paper-page-sm.svg)](https://huggingface.co/papers/2401.16468) -[Marcos V. Conde](https://scholar.google.com/citations?user=NtB1kjYAAAAJ&hl=en), [Gregor Geigle](https://scholar.google.com/citations?user=uIlyqRwAAAAJ&hl=en), [Radu Timofte](https://scholar.google.com/citations?user=u3MwH5kAAAAJ&hl=en) +[Marcos V. Conde](https://mv-lab.github.io/), [Gregor Geigle](https://scholar.google.com/citations?user=uIlyqRwAAAAJ&hl=en), [Radu Timofte](https://scholar.google.com/citations?user=u3MwH5kAAAAJ&hl=en) Computer Vision Lab, University of Wuerzburg | Sony PlayStation, FTG -InstructIR + +InstructIR + +Video courtesy of Gradio ([see their post about InstructIR](https://twitter.com/Gradio/status/1752776176811041049)). Also shoutout to AK -- [see his tweet](https://twitter.com/_akhaliq/status/1752551364566126798). + ### TL;DR: quickstart InstructIR takes as input an image and a human-written instruction for how to improve that image. The neural model performs all-in-one image restoration. InstructIR achieves state-of-the-art results on several restoration tasks including image denoising, deraining, deblurring, dehazing, and (low-light) image enhancement. @@ -45,12 +49,9 @@ Image restoration is a fundamental problem that involves recovering a high-quali 🚀 You can start with the [demo tutorial](demo.ipynb). We also host the same tutorial on [google colab](https://colab.research.google.com/drive/1OrTvS-i6uLM2Y8kIkq8ZZRwEQxQFchfq?usp=sharing) so you can run it using free GPUs!. -| | | -|----------|:-------------: -| InstructIR App | InstructIR App | - +InstructIR -### Gradio Demo +### Gradio Demo We made a simple [Gradio demo](app.py) you can run (locally) on your machine [here](app.py). You need Python>=3.9 and [these requirements](requirements_gradio.txt) for it: `pip install -r requirements_gradio.txt` ``` diff --git a/images/instructir.gif b/images/instructir.gif new file mode 100644 index 0000000..610d651 Binary files /dev/null and b/images/instructir.gif differ diff --git a/images/instructir.mp4 b/images/instructir.mp4 new file mode 100644 index 0000000..7b4d164 Binary files /dev/null and b/images/instructir.mp4 differ diff --git a/index.html b/index.html index 658dda2..63e79d2 100644 --- a/index.html +++ b/index.html @@ -117,7 +117,7 @@

InstructIR: High-Quality Image Restorat
- Marcos V. Conde1,2, + Marcos V. Conde1,2, Gregor Geigle1, @@ -195,7 +195,7 @@

InstructIR: High-Quality Image Restorat
-

TL;DR & Abstract

+

TL;DR

@@ -211,16 +211,6 @@

TL;DR & Abstract



-
-
-

Abstract

(click me to read) -

- Image restoration is a fundamental problem that involves recovering a high-quality clean image from its degraded observation. All-In-One image restoration models can effectively restore images from various types and levels of degradation using degradation-specific information as prompts to guide the restoration model. In this work, we present the first approach that uses human-written instructions to guide the image restoration model. Given natural language prompts, our model can recover high-quality images from their degraded counterparts, considering multiple degradation types. Our method, InstructIR, achieves state-of-the-art results on several restoration tasks including image denoising, deraining, deblurring, dehazing, and (low-light) image enhancement. InstructIR improves +1dB over previous all-in-one restoration methods. Moreover, our dataset and results represent a novel benchmark for new research on text-guided image restoration and enhancement. -

-
-
- -
@@ -236,7 +226,17 @@

TL;DR & Abstract

Examples of InstructIR

- + + + + + +