https://github.com/daxrajsinh/image-quality-assessment_ml-dl
"An Advanced ML-DL based project for evaluating and enhancing image aesthetics resulting "Aesthetics score" using CLIP IQA, CLIP , and BLIP models." Provides customizable prompts to the users.
https://github.com/daxrajsinh/image-quality-assessment_ml-dl
blip-classification clip-classification clip-iqa deep-learning machine-learning
Last synced: 3 months ago
JSON representation
"An Advanced ML-DL based project for evaluating and enhancing image aesthetics resulting "Aesthetics score" using CLIP IQA, CLIP , and BLIP models." Provides customizable prompts to the users.
- Host: GitHub
- URL: https://github.com/daxrajsinh/image-quality-assessment_ml-dl
- Owner: Daxrajsinh
- Created: 2024-06-15T18:56:37.000Z (12 months ago)
- Default Branch: main
- Last Pushed: 2024-06-15T19:56:20.000Z (12 months ago)
- Last Synced: 2024-06-15T20:45:44.637Z (12 months ago)
- Topics: blip-classification, clip-classification, clip-iqa, deep-learning, machine-learning
- Language: Jupyter Notebook
- Homepage:
- Size: 7.81 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Welcome to the **Image Quality Assessment**, An ML, DL based project! This repository hosts a cutting-edge tool for evaluating the aesthetic quality of images (By Providing "Quality of an image" score of USER INPUT image) using advanced AI, ML, and deep learning techniques. Whether you are a photographer, designer, or an enthusiast of visual aesthetics, this empowers you to assess, select and enhance the quality of your images effortlessly.
## 🚀 Features
- **🔍 Multimodal Analysis**: Leverages both visual and textual data for comprehensive image quality assessment.
- **🎨 Customizable Prompts**: Allows users to define personalized criteria for evaluating image aesthetics.
- **🤖 Advanced AI Models**: Utilizes CLIP IQA (Image Quality Assessment), CLIP:Classification, and BLIP:Classification for precise and context-aware image evaluations.
## 📦 Installation
1. **Clone the Repository**:
```bash
git clone https://github.com/yourusername/Image_Quality_Assessment.git
cd Image_Quality_Assessment
2. **Install Dependencies:**:
```bash
pip install torch torchvision pillow lavis pandas
## 🛠️ Usage
You can upload your images to the repository and use the provided functions to evaluate their aesthetic scores. Follow these steps:
1. **Step 1: Load and Process Images**: Place your images in the `images` folder.
2. **Step 2: Customize Prompts**: Modify the `cls_names` list to use your own descriptions for good(Postives) and bad(Negatives) photos.
3. **Step 3: Run the Evaluation Script**: Modify the `img_files` list in the evaluation script to include the paths to your uploaded images.
```python
img_files = ['images/your_image1.jpg', 'images/your_image2.jpg']
## 🎨 Examples
Evaluating a Single Image
```python
from PIL import Image
img = Image.open("path/to/your/image.jpg")
aesthetics_score = score_image(img)
print(f"Aesthetics Score: {aesthetics_score}")
```
Batch Evaluation
```python
img_files = ["path/to/your/image1.jpg", "path/to/your/image2.jpg"]
scores = []
for img_file in img_files:
img = Image.open(img_file)
aesthetics_score = score_image(img)
scores.append({'img': img_file.split('/')[-1], 'score': aesthetics_score})
print(f"{img_file.split('/')[-1]} - Aesthetics Score: {aesthetics_score}")
import pandas as pd
pd.DataFrame(scores).describe()
```
## 🤝 Contributing
We welcome contributions to enhance this project! Feel free to submit issues or pull requests on GitHub.