Spaces:
Running
Running
feat:added readme.md
Browse files- README.md +15 -145
- READMEs.md +152 -0
- readme.md +0 -22
README.md
CHANGED
@@ -1,152 +1,22 @@
|
|
1 |
-
# AI-Contain-Checker
|
2 |
-
|
3 |
-
A modular AI content detection system with support for **image classification**, **image edit detection**, **Nepali text classification**, and **general text classification**. Built for performance and extensibility, it is ideal for detecting AI-generated content in both visual and textual forms.
|
4 |
-
|
5 |
-
|
6 |
-
## π Features
|
7 |
-
|
8 |
-
### πΌοΈ Image Classifier
|
9 |
-
|
10 |
-
* **Purpose**: Classifies whether an image is AI-generated or a real-life photo.
|
11 |
-
* **Model**: Fine-tuned **InceptionV3** CNN.
|
12 |
-
* **Dataset**: Custom curated dataset with **\~79,950 images** for binary classification.
|
13 |
-
* **Location**: [`features/image_classifier`](features/image_classifier)
|
14 |
-
* **Docs**: [`docs/features/image_classifier.md`](docs/features/image_classifier.md)
|
15 |
-
|
16 |
-
### ποΈ Image Edit Detector
|
17 |
-
|
18 |
-
* **Purpose**: Detects image tampering or post-processing.
|
19 |
-
* **Techniques Used**:
|
20 |
-
|
21 |
-
* **Error Level Analysis (ELA)**: Visualizes compression artifacts.
|
22 |
-
* **Fast Fourier Transform (FFT)**: Detects unnatural frequency patterns.
|
23 |
-
* **Location**: [`features/image_edit_detector`](features/image_edit_detector)
|
24 |
-
* **Docs**:
|
25 |
-
|
26 |
-
* [ELA](docs/detector/ELA.md)
|
27 |
-
* [FFT](docs/detector/fft.md )
|
28 |
-
* [Metadata Analysis](docs/detector/meta.md)
|
29 |
-
* [Backend Notes](docs/detector/note-for-backend.md)
|
30 |
-
|
31 |
-
### π Nepali Text Classifier
|
32 |
-
|
33 |
-
* **Purpose**: Determines if Nepali text content is AI-generated or written by a human.
|
34 |
-
* **Model**: Based on `XLMRClassifier` fine-tuned on Nepali language data.
|
35 |
-
* **Dataset**: Scraped dataset of **\~18,000** Nepali texts.
|
36 |
-
* **Location**: [`features/nepali_text_classifier`](features/nepali_text_classifier)
|
37 |
-
* **Docs**: [`docs/features/nepali_text_classifier.md`](docs/features/nepali_text_classifier.md)
|
38 |
-
|
39 |
-
### π English Text Classifier
|
40 |
-
|
41 |
-
* **Purpose**: Detects if English text is AI-generated or human-written.
|
42 |
-
* **Pipeline**:
|
43 |
-
|
44 |
-
* Uses **GPT2 tokenizer** for input preprocessing.
|
45 |
-
* Custom binary classifier to differentiate between AI and human-written content.
|
46 |
-
* **Location**: [`features/text_classifier`](features/text_classifier)
|
47 |
-
* **Docs**: [`docs/features/text_classifier.md`](docs/features/text_classifier.md)
|
48 |
-
|
49 |
---
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
βββ Dockerfile # Docker build script
|
59 |
-
βββ Procfile # Deployment file for Heroku or similar
|
60 |
-
βββ requirements.txt # Python dependencies
|
61 |
-
βββ README.md # You are here π
|
62 |
-
β
|
63 |
-
βββ features/ # Core detection modules
|
64 |
-
β βββ image_classifier/
|
65 |
-
β βββ image_edit_detector/
|
66 |
-
β βββ nepali_text_classifier/
|
67 |
-
β βββ text_classifier/
|
68 |
-
β
|
69 |
-
βββ docs/ # Internal and API documentation
|
70 |
-
β βββ api_endpoints.md
|
71 |
-
β βββ deployment.md
|
72 |
-
β βββ detector/
|
73 |
-
β β βββ ELA.md
|
74 |
-
β β βββ fft.md
|
75 |
-
β β βββ meta.md
|
76 |
-
β β βββ note-for-backend.md
|
77 |
-
β βββ functions.md
|
78 |
-
β βββ nestjs_integration.md
|
79 |
-
β βββ security.md
|
80 |
-
β βββ setup.md
|
81 |
-
β βββ structure.md
|
82 |
-
β
|
83 |
-
βββ IMG_Models/ # Saved image classifier model(s)
|
84 |
-
β βββ latest-my_cnn_model.h5
|
85 |
-
β
|
86 |
-
βββ notebooks/ # Experimental and debug notebooks
|
87 |
-
βββ static/ # Static assets if needed
|
88 |
-
βββ test.md # Test notes
|
89 |
-
````
|
90 |
-
|
91 |
---
|
92 |
|
93 |
-
|
94 |
|
95 |
-
|
96 |
-
* [Deployment Guide](docs/deployment.md)
|
97 |
-
* [Detector Documentation](docs/detector/)
|
98 |
|
99 |
-
|
100 |
-
* [Fast Fourier Transform (FFT)](docs/detector/fft.md)
|
101 |
-
* [Metadata Analysis](docs/detector/meta.md)
|
102 |
-
* [Backend Notes](docs/detector/note-for-backend.md)
|
103 |
-
* [Functions Overview](docs/functions.md)
|
104 |
-
* [NestJS Integration Guide](docs/nestjs_integration.md)
|
105 |
-
* [Security Details](docs/security.md)
|
106 |
-
* [Setup Instructions](docs/setup.md)
|
107 |
-
* [Project Structure](docs/structure.md)
|
108 |
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
-
1. **Install dependencies**
|
114 |
-
|
115 |
-
```bash
|
116 |
-
pip install -r requirements.txt
|
117 |
-
```
|
118 |
-
|
119 |
-
2. **Run the API**
|
120 |
-
|
121 |
-
```bash
|
122 |
-
uvicorn app:app --reload
|
123 |
-
```
|
124 |
-
|
125 |
-
3. **Build Docker (optional)**
|
126 |
-
|
127 |
-
```bash
|
128 |
-
docker build -t ai-contain-checker .
|
129 |
-
docker run -p 8000:8000 ai-contain-checker
|
130 |
-
```
|
131 |
-
|
132 |
-
---
|
133 |
-
|
134 |
-
## π Security & Integration
|
135 |
-
|
136 |
-
* **Token Authentication** and **IP Whitelisting** supported.
|
137 |
-
* NestJS integration guide: [`docs/nestjs_integration.md`](docs/nestjs_integration.md)
|
138 |
-
* Rate limiting handled using `slowapi`.
|
139 |
-
|
140 |
-
---
|
141 |
-
|
142 |
-
## π‘οΈ Future Plans
|
143 |
-
|
144 |
-
* Add **video classifier** module.
|
145 |
-
* Expand dataset for **multilingual** AI content detection.
|
146 |
-
* Add **fine-tuning UI** for models.
|
147 |
-
|
148 |
-
---
|
149 |
-
|
150 |
-
## π License
|
151 |
|
152 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
title: Testing AI Contain
|
3 |
+
emoji: π€
|
4 |
+
colorFrom: blue
|
5 |
+
colorTo: green
|
6 |
+
sdk: docker
|
7 |
+
sdk_version: "latest"
|
8 |
+
app_file: app.py
|
9 |
+
pinned: false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
---
|
11 |
|
12 |
+
# Testing AI Contain
|
13 |
|
14 |
+
This Hugging Face Space uses **Docker** to run a custom environment for AI content detection.
|
|
|
|
|
15 |
|
16 |
+
## How to run locally
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
|
18 |
+
```bash
|
19 |
+
docker build -t testing-ai-contain .
|
20 |
+
docker run -p 7860:7860 testing-ai-contain
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
|
22 |
+
```
|
READMEs.md
ADDED
@@ -0,0 +1,152 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# AI-Contain-Checker
|
2 |
+
|
3 |
+
A modular AI content detection system with support for **image classification**, **image edit detection**, **Nepali text classification**, and **general text classification**. Built for performance and extensibility, it is ideal for detecting AI-generated content in both visual and textual forms.
|
4 |
+
|
5 |
+
|
6 |
+
## π Features
|
7 |
+
|
8 |
+
### πΌοΈ Image Classifier
|
9 |
+
|
10 |
+
* **Purpose**: Classifies whether an image is AI-generated or a real-life photo.
|
11 |
+
* **Model**: Fine-tuned **InceptionV3** CNN.
|
12 |
+
* **Dataset**: Custom curated dataset with **\~79,950 images** for binary classification.
|
13 |
+
* **Location**: [`features/image_classifier`](features/image_classifier)
|
14 |
+
* **Docs**: [`docs/features/image_classifier.md`](docs/features/image_classifier.md)
|
15 |
+
|
16 |
+
### ποΈ Image Edit Detector
|
17 |
+
|
18 |
+
* **Purpose**: Detects image tampering or post-processing.
|
19 |
+
* **Techniques Used**:
|
20 |
+
|
21 |
+
* **Error Level Analysis (ELA)**: Visualizes compression artifacts.
|
22 |
+
* **Fast Fourier Transform (FFT)**: Detects unnatural frequency patterns.
|
23 |
+
* **Location**: [`features/image_edit_detector`](features/image_edit_detector)
|
24 |
+
* **Docs**:
|
25 |
+
|
26 |
+
* [ELA](docs/detector/ELA.md)
|
27 |
+
* [FFT](docs/detector/fft.md )
|
28 |
+
* [Metadata Analysis](docs/detector/meta.md)
|
29 |
+
* [Backend Notes](docs/detector/note-for-backend.md)
|
30 |
+
|
31 |
+
### π Nepali Text Classifier
|
32 |
+
|
33 |
+
* **Purpose**: Determines if Nepali text content is AI-generated or written by a human.
|
34 |
+
* **Model**: Based on `XLMRClassifier` fine-tuned on Nepali language data.
|
35 |
+
* **Dataset**: Scraped dataset of **\~18,000** Nepali texts.
|
36 |
+
* **Location**: [`features/nepali_text_classifier`](features/nepali_text_classifier)
|
37 |
+
* **Docs**: [`docs/features/nepali_text_classifier.md`](docs/features/nepali_text_classifier.md)
|
38 |
+
|
39 |
+
### π English Text Classifier
|
40 |
+
|
41 |
+
* **Purpose**: Detects if English text is AI-generated or human-written.
|
42 |
+
* **Pipeline**:
|
43 |
+
|
44 |
+
* Uses **GPT2 tokenizer** for input preprocessing.
|
45 |
+
* Custom binary classifier to differentiate between AI and human-written content.
|
46 |
+
* **Location**: [`features/text_classifier`](features/text_classifier)
|
47 |
+
* **Docs**: [`docs/features/text_classifier.md`](docs/features/text_classifier.md)
|
48 |
+
|
49 |
+
---
|
50 |
+
|
51 |
+
## ποΈ Project Structure
|
52 |
+
|
53 |
+
```bash
|
54 |
+
AI-Checker/
|
55 |
+
β
|
56 |
+
βββ app.py # Main FastAPI entry point
|
57 |
+
βββ config.py # Configuration settings
|
58 |
+
βββ Dockerfile # Docker build script
|
59 |
+
βββ Procfile # Deployment file for Heroku or similar
|
60 |
+
βββ requirements.txt # Python dependencies
|
61 |
+
βββ README.md # You are here π
|
62 |
+
β
|
63 |
+
βββ features/ # Core detection modules
|
64 |
+
β βββ image_classifier/
|
65 |
+
β βββ image_edit_detector/
|
66 |
+
β βββ nepali_text_classifier/
|
67 |
+
β βββ text_classifier/
|
68 |
+
β
|
69 |
+
βββ docs/ # Internal and API documentation
|
70 |
+
β βββ api_endpoints.md
|
71 |
+
β βββ deployment.md
|
72 |
+
β βββ detector/
|
73 |
+
β β βββ ELA.md
|
74 |
+
β β βββ fft.md
|
75 |
+
β β βββ meta.md
|
76 |
+
β β βββ note-for-backend.md
|
77 |
+
β βββ functions.md
|
78 |
+
β βββ nestjs_integration.md
|
79 |
+
β βββ security.md
|
80 |
+
β βββ setup.md
|
81 |
+
β βββ structure.md
|
82 |
+
β
|
83 |
+
βββ IMG_Models/ # Saved image classifier model(s)
|
84 |
+
β βββ latest-my_cnn_model.h5
|
85 |
+
β
|
86 |
+
βββ notebooks/ # Experimental and debug notebooks
|
87 |
+
βββ static/ # Static assets if needed
|
88 |
+
βββ test.md # Test notes
|
89 |
+
````
|
90 |
+
|
91 |
+
---
|
92 |
+
|
93 |
+
## π Documentation Links
|
94 |
+
|
95 |
+
* [API Endpoints](docs/api_endpoints.md)
|
96 |
+
* [Deployment Guide](docs/deployment.md)
|
97 |
+
* [Detector Documentation](docs/detector/)
|
98 |
+
|
99 |
+
* [Error Level Analysis (ELA)](docs/detector/ELA.md)
|
100 |
+
* [Fast Fourier Transform (FFT)](docs/detector/fft.md)
|
101 |
+
* [Metadata Analysis](docs/detector/meta.md)
|
102 |
+
* [Backend Notes](docs/detector/note-for-backend.md)
|
103 |
+
* [Functions Overview](docs/functions.md)
|
104 |
+
* [NestJS Integration Guide](docs/nestjs_integration.md)
|
105 |
+
* [Security Details](docs/security.md)
|
106 |
+
* [Setup Instructions](docs/setup.md)
|
107 |
+
* [Project Structure](docs/structure.md)
|
108 |
+
|
109 |
+
---
|
110 |
+
|
111 |
+
## π Usage
|
112 |
+
|
113 |
+
1. **Install dependencies**
|
114 |
+
|
115 |
+
```bash
|
116 |
+
pip install -r requirements.txt
|
117 |
+
```
|
118 |
+
|
119 |
+
2. **Run the API**
|
120 |
+
|
121 |
+
```bash
|
122 |
+
uvicorn app:app --reload
|
123 |
+
```
|
124 |
+
|
125 |
+
3. **Build Docker (optional)**
|
126 |
+
|
127 |
+
```bash
|
128 |
+
docker build -t ai-contain-checker .
|
129 |
+
docker run -p 8000:8000 ai-contain-checker
|
130 |
+
```
|
131 |
+
|
132 |
+
---
|
133 |
+
|
134 |
+
## π Security & Integration
|
135 |
+
|
136 |
+
* **Token Authentication** and **IP Whitelisting** supported.
|
137 |
+
* NestJS integration guide: [`docs/nestjs_integration.md`](docs/nestjs_integration.md)
|
138 |
+
* Rate limiting handled using `slowapi`.
|
139 |
+
|
140 |
+
---
|
141 |
+
|
142 |
+
## π‘οΈ Future Plans
|
143 |
+
|
144 |
+
* Add **video classifier** module.
|
145 |
+
* Expand dataset for **multilingual** AI content detection.
|
146 |
+
* Add **fine-tuning UI** for models.
|
147 |
+
|
148 |
+
---
|
149 |
+
|
150 |
+
## π License
|
151 |
+
|
152 |
+
See full license terms here: [`LICENSE.md`](license.md)
|
readme.md
DELETED
@@ -1,22 +0,0 @@
|
|
1 |
-
---
|
2 |
-
title: Testing AI Contain
|
3 |
-
emoji: π€
|
4 |
-
colorFrom: blue
|
5 |
-
colorTo: green
|
6 |
-
sdk: docker
|
7 |
-
sdk_version: "latest"
|
8 |
-
app_file: app.py
|
9 |
-
pinned: false
|
10 |
-
---
|
11 |
-
|
12 |
-
# Testing AI Contain
|
13 |
-
|
14 |
-
This Hugging Face Space uses **Docker** to run a custom environment for AI content detection.
|
15 |
-
|
16 |
-
## How to run locally
|
17 |
-
|
18 |
-
```bash
|
19 |
-
docker build -t testing-ai-contain .
|
20 |
-
docker run -p 7860:7860 testing-ai-contain
|
21 |
-
|
22 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|