🎨 Cartoonify Studio Pro

Enterprise AI & Computer Vision Image Cartoonization Suite

Transform ordinary photos, portraits, and landscapes into stunning anime, Studio Ghibli, pop-art comic, and watercolor artwork. Features an ultra-modern Web Studio with an interactive Before/After split slider, a sleek Desktop GUI, a developer-first CLI & Python SDK, and a static Showcase Landing Page.

🌐 Live Showcase Website • ✨ Web Studio • 🖥️ Desktop GUI • ⚡ Python SDK • 🛡️ Security Matrix

---

🌟 13 Curated Visual Styles

Style	Category	Description	Accent
🔮 Studio Ghibli (Generative AI)	Generative AI	ChatGPT (DALL-E 3) & SDXL generative synthesis: reimagines scene with authentic Miyazaki character design.	#E84393
🎬 Studio Ghibli Pro	Artistic CV	Kuwahara painterly brushwork with warm afternoon sunlight grading and anti-aliased pencil inking.	#FF6B6B
🍃 Hayao Anime	Neural AI	PyTorch AnimeGANv2 trained directly on Miyazaki anime frames.	#2ECC71
✨ Anime Portrait Soft	Neural AI	State-of-the-art neural anime stylization with expressive eyes and crisp face preservation.	#4ECDC4
🎨 Paprika Vibrant	Neural AI	Warm saturated cinematic cartoon look inspired by Satoshi Kon feature animation.	#FFB84D
🌌 Shinkai Luminous	Neural AI	Makoto Shinkai high-contrast radiant lighting and glowing atmospheric accents.	#9B59B6
💥 Comic Pop Art	Artistic CV	Bold graphic novel inked contours with vibrant cell-shaded primary colors.	#E74C3C
🎭 Watercolor Dream	Artistic CV	Soft pigment diffusion, gentle paper texture wash, and fluid color bleeding.	#A8D8EA
✏️ Pencil & Charcoal	Artistic CV	Detailed graphite crosshatch line work with smooth continuous shading.	#95A5A6
🖍️ Colored Pencil	Artistic CV	Hand-drawn color sketch with textured paper grain and vibrant strokes.	#F39C12
⚡ Neon Cyberpunk	Artistic CV	Darkened cinematic scene with electric cyan and magenta glowing edges.	#00F0FF
📺 Retro 90s TV Toon	Artistic CV	Nostalgic Saturday morning cartoon aesthetic with thick black ink borders.	#FD79A8
🎛️ Custom Pro Shader	Parametric	Real-time tunable parameters: line thickness, opacity, color count, saturation, and contrast.	#00CEC9

---

🚀 Key Features

• 🌐 Modern Web Studio: Glassmorphism UI with interactive drag-and-drop, real-time Before/After comparison slider, live camera snapshot, and responsive dark/light themes.
• ✨ GitHub Pages Showcase (docs/): Zero-dependency static landing page with an in-browser live HTML5 Canvas cartoon simulator.
• 🖥️ Desktop GUI: Cross-platform responsive desktop app with side-by-side comparison, non-blocking asynchronous processing threads, and history gallery.
• ⚡ Neural & CV Engine: Combines deep learning (PyTorch Hub AnimeGANv2) and high-performance OpenCV algorithms for instant, offline-capable rendering.
• 👤 Face-Preserved Alignment: Smart face detection with elliptical feathered alpha masking to preserve identity on portrait photos.
• 📦 Batch Processing: Convert entire image directories in seconds and export directly as a ZIP archive.
• 🛡️ 50/50 Security & Fuzz Tests: Protected against decompression bombs, path traversal, Zip Slip, and NaN/Inf boundary vectors.
• 🐳 Docker Ready: Containerized deployment via Dockerfile and Docker Compose.

---

🔬 Algorithmic Blueprint & Math Models

graph TD
    A[Input: File / Webcam / Base64 / Stream] --> B[Security Validator & Clamping Layer]
    B --> C{Face Alignment Enabled?}
    C -->|Yes| D[OpenCV Haar Cascade Face Detector]
    D --> E[Proportional 1.6x ROI Expansion]
    E --> F[Generate Feathered Elliptical Alpha Mask]
    C -->|No| G[Full Frame Ingestion]
    
    F --> H{Selected Style Category}
    G --> H
    
    H -->|Neural AI| I[AnimeGANv2 Generator: face_paint / paprika / hayao / shinkai]
    H -->|Artistic CV| J[Ghibli Pro / Pop Art / Watercolor / Neon / Pencil]
    H -->|Parametric| K[Custom Shader: Line Inking + K-Means Color Quantization]
    
    I --> L[Post-Processing & Local Contrast CLAHE]
    J --> L
    K --> L
    
    L --> M{Face Composite Needed?}
    M -->|Yes| N[Feathered Alpha Blend: Styled Face + Original Backdrop]
    M -->|No| O[Final Output Image]
    N --> O

Loading

1. Multi-Pass Bilateral Color Smoothing (Studio Ghibli Pro)

Smooths flat surfaces while preserving sharp edge boundaries: $$I_{\text{smooth}}(x) = \frac{1}{W_p} \sum_{x_i \in \Omega} I(x_i) \cdot g_s(|x_i - x|) \cdot g_r(|I(x_i) - I(x)|)$$

2. Feathered Gaussian Alpha Mask (Face Preservation)

Calculates a 1.6x expanded centroid bounding box around detected facial coordinates and applies a continuous Gaussian fade to eliminate square seams: $$\alpha(x, y) = \exp\left(-\left(\frac{(x - c_x)^2}{2\sigma_x^2} + \frac{(y - c_y)^2}{2\sigma_y^2}\right)\right)$$

3. Color Dodge Sketch Blending (Pencil & Charcoal)

$$I_{\text{sketch}}(x, y) = \min\left(255, \frac{I_{\text{gray}}(x, y) \cdot 256}{255 - I_{\text{inv_blur}}(x, y) + 1}\right)$$

4. K-Means Cell Quantization (Comic Pop Art)

$$\arg\min_{\mu} \sum_{i=1}^{N} \min_{k} |x_i - \mu_k|^2$$

---

🛡️ Enterprise Security Hardening & Fuzz Testing

The codebase includes an enterprise security test suite in tests/test_security.py:

Protection Layer	Security Invariant	Status
Pixel Decompression Bomb	Pillow 50M limit + 40MP max dimension rejection	✅ PASSED
Path Traversal Defense	Strips ../../, ..\Windows, null bytes \x00, special characters	✅ PASSED
Zip Slip & Zip Bomb	Capped at 50 files / 200MB uncompressed buffer	✅ PASSED
Adversarial Fuzzing	Handles NaN, +Infinity, -Infinity, negative slider values	✅ PASSED
Corrupted Stream Resilience	Handles mutated bit-flipped byte streams and invalid Base64	✅ PASSED
HTTP Security Headers	X-Content-Type-Options: nosniff, X-Frame-Options: SAMEORIGIN	✅ PASSED
Thread Safety	Mutex-guarded neural model cache with zero race conditions	✅ PASSED

# Run the complete test suite:
pytest tests/ -v
# ===================== 50 passed in 56.43s =====================

---

🛠️ Quick Installation

1. Clone the Repository

git clone https://github.com/AryanXCode646/cartoon-image-generator67.git
cd cartoon-image-generator67

2. Create & Activate Virtual Environment

python -m venv .venv

# Windows (PowerShell)
.\.venv\Scripts\Activate.ps1

# Linux / macOS
source .venv/bin/activate

3. Install Dependencies

pip install -r requirements.txt
# Or install in editable developer mode:
pip install -e .

---

🎯 How to Run

🌐 1. Launch the Web Studio (Recommended)

# Windows 1-click launcher:
run_web.bat

# Or via CLI:
python -m cartoonify web

Then open your browser at http://127.0.0.1:8000.

---

🖥️ 2. Launch the Desktop GUI

# Windows 1-click launcher:
run_gui.bat

# Or via CLI:
python -m cartoonify gui

---

💻 3. Command-Line Interface (CLI)

Process a single image:

python -m cartoonify process samples/sample_portrait.jpg -o output_ghibli.jpg --style ghibli_pro --strength 0.85

Process with face-aligned preservation:

python -m cartoonify process selfie.jpg -o selfie_anime.jpg --style anime_soft --face-align

Batch process an entire directory:

python -m cartoonify batch ./my_vacation_photos -o ./cartoons --style comic_pop

List all available styles:

python -m cartoonify list-styles

---

🐍 4. Python SDK Integration

from cartoonify import CartoonEngine
import cv2

# Initialize the engine (auto-selects CUDA GPU or CPU)
engine = CartoonEngine()

# 1. ChatGPT-Grade Generative AI Studio Ghibli Synthesis
engine.process_file(
    "portrait.jpg",
    "ghibli_chatgpt_art.jpg",
    style="ghibli_generative",
    custom_params={"openai_api_key": "your-openai-api-key"}
)

# 2. Local Neural Hayao Miyazaki Checkpoint (AnimeGANv2)
engine.process_file(
    "portrait.jpg",
    "ghibli_hayao_art.jpg",
    style="hayao",
    use_face_align=True
)

# 3. High-Speed Kuwahara Painterly Filter (100% Offline)
engine.process_file(
    "portrait.jpg",
    "ghibli_kuwahara.jpg",
    style="ghibli_pro",
    strength=0.85
)

---

🐳 5. Docker Deployment

# Start Web Studio in container:
docker-compose up -d

# Open in browser:
http://localhost:8000

---

⚡ Windows 1-Click Launchers

Launcher Script	Description
run_web.bat	Launches FastAPI server & opens Web Studio in default browser
run_gui.bat	Launches the modern Tkinter Desktop GUI
run_cli.bat	Opens interactive CLI helper
run_tests.bat	Executes the 50/50 Pytest suite
run_showcase_website.bat	Opens the static showcase website locally

---

🧪 Testing & CI/CD

# Run all 50 unit and security tests:
pytest tests/ -v

# Run with coverage:
pytest --cov=cartoonify tests/

Automated GitHub Actions CI runs on every push across Ubuntu (Linux), Windows, and macOS.

---

📄 License

Distributed under the MIT License. See LICENSE for more details.

---

Built with ❤️ by Aryan • Open Source AI & Computer Vision