Spaces:
Configuration error
Configuration error
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,58 +1,64 @@
|
|
| 1 |
-
|
|
|
|
|
|
|
| 2 |
|
| 3 |
Description
|
| 4 |
|
| 5 |
-
|
|
|
|
|
|
|
| 6 |
|
| 7 |
-
|
|
|
|
|
|
|
| 8 |
|
| 9 |
-
Génération automatique d'images à partir d'une image source.
|
| 10 |
-
Possibilité de contrôler différents aspects de la génération, tels que la résolution, le style, la couleur, etc.
|
| 11 |
-
Interface utilisateur intuitive pour faciliter l'utilisation.
|
| 12 |
Installation
|
| 13 |
|
| 14 |
-
|
| 15 |
bash
|
| 16 |
-
Copy code
|
| 17 |
-
```git clone https://huggingface.co/spaces/Antropophobe/QR_Code```
|
| 18 |
-
Assurez-vous d'avoir Python installé sur votre système. Ce projet utilise Python 3.
|
| 19 |
-
Installez les dépendances requises en exécutant la commande suivante :
|
| 20 |
|
| 21 |
-
|
| 22 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
css
|
| 24 |
|
| 25 |
-
|
| 26 |
-
|
|
|
|
|
|
|
| 27 |
|
| 28 |
-
|
| 29 |
-
- Télécharger le modèle de génération d'image que vous souhaitez, j'ai personnelement utilisé GhostMix (https://civitai.com/models/36520/ghostmix)
|
| 30 |
-
- Ajouter le model dans le fichier .../models/Stabble-diffusion/
|
| 31 |
|
| 32 |
-
|
| 33 |
-
- Ajouter les models dans le fichier ../models/controlsNet
|
| 34 |
|
|
|
|
| 35 |
|
| 36 |
-
|
| 37 |
-
|
|
|
|
|
|
|
|
|
|
| 38 |
Method : txt2Img
|
| 39 |
-
|
| 40 |
-
Width : 768
|
| 41 |
Width : 768
|
| 42 |
-
|
| 43 |
-
|
|
|
|
| 44 |
Enable
|
| 45 |
-
(
|
| 46 |
-
Pixel
|
| 47 |
-
Control
|
| 48 |
Preprocessor : all
|
| 49 |
Model : QR_monster / Tile_ControlNet
|
| 50 |
-
Control Weight : 1.7 (
|
| 51 |
Starting Control Step : 0.2
|
| 52 |
-
Ending Control Step : 0.8-0.9 (
|
| 53 |
-
|
| 54 |
Contribution
|
| 55 |
-
--
|
| 56 |
-
Prompt : ce que vous souhaitez obtenir.
|
| 57 |
|
| 58 |
-
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
|
| 3 |
+
Title: AI Project: Generating Images From a Given Input Image
|
| 4 |
|
| 5 |
Description
|
| 6 |
|
| 7 |
+
This artificial intelligence project aims to generate unique images based on a given input image. It employs image processing and deep learning techniques to understand the structure and characteristics of the input image and produce new images that share similarities with it.
|
| 8 |
+
|
| 9 |
+
Features
|
| 10 |
|
| 11 |
+
Automatically generate images from a given input image.
|
| 12 |
+
Ability to control various aspects of generation such as resolution, style, color, etc.
|
| 13 |
+
User-friendly interface for ease of use.
|
| 14 |
|
|
|
|
|
|
|
|
|
|
| 15 |
Installation
|
| 16 |
|
| 17 |
+
Clone this repository onto your local machine:
|
| 18 |
bash
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
|
| 20 |
+
git clone https://huggingface.co/spaces/Antropophobe/QR_Code
|
| 21 |
+
Ensure you have Python installed on your system. This project uses Python 3.
|
| 22 |
+
Install required dependencies by executing the following command:
|
| 23 |
+
|
| 24 |
+
Place your input image in the dedicated folder within the project directory.
|
| 25 |
+
Launch the main program via the following command:
|
| 26 |
css
|
| 27 |
|
| 28 |
+
./webui.sh Mac
|
| 29 |
+
./websui.bat Windows
|
| 30 |
+
|
| 31 |
+
Steps to follow:
|
| 32 |
|
| 33 |
+
Install the ControlNet extension: <https://github.com/lllyasviel/ControlNet>
|
|
|
|
|
|
|
| 34 |
|
| 35 |
+
Download the desired image generation model, I personally recommend GhostMix (<https://civitai.com/models/36520/ghostmix>)
|
|
|
|
| 36 |
|
| 37 |
+
Add the model to the file located at .../models/Stabble-diffusion/
|
| 38 |
|
| 39 |
+
Download ControlNet models (<https://github.com/lllyasviel/ControlNet-v1-1-nightly/tree/main/models>)
|
| 40 |
+
|
| 41 |
+
Add the models to the file located at ../models/controlsNet
|
| 42 |
+
|
| 43 |
+
Presets for generating a QR code:
|
| 44 |
Method : txt2Img
|
| 45 |
+
Sampling Steps : 30-50
|
|
|
|
| 46 |
Width : 768
|
| 47 |
+
Height : 768
|
| 48 |
+
|
| 49 |
+
ControlNet Parameters:
|
| 50 |
Enable
|
| 51 |
+
(Add QR code - generatable at <https://fr.qr-code-generator.com>)
|
| 52 |
+
Pixel Perfect
|
| 53 |
+
Control Type : All
|
| 54 |
Preprocessor : all
|
| 55 |
Model : QR_monster / Tile_ControlNet
|
| 56 |
+
Control Weight : 1.7 (the higher, the more visible the QR code will be in the result)
|
| 57 |
Starting Control Step : 0.2
|
| 58 |
+
Ending Control Step : 0.8-0.9 (depending on the result)
|
| 59 |
+
Remember to save presets.
|
| 60 |
Contribution
|
|
|
|
|
|
|
| 61 |
|
| 62 |
+
Prompt : What you wish to obtain.
|
| 63 |
+
|
| 64 |
+
Negative Prompt : ", blur, ugliness, poor artistic quality, visible signatures, low contrast, underexposure, amateurish appearance, distortion, and multiple angles. Ensure that the final image meets the highest standards of clarity, beauty, and professional craftsmanship."
|