Skip to content

Latest commit

 

History

32 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DivyaLipi-AI

Sanskrit OCR and Translation of degraded Sanskrit manuscripts using Deep Learning


📽️ Demo

Check out the live demo and results here:
🔗 LinkedIn Demo Post


🚧 Commit Structure Notice

Due to computational and GPU hardware limitations, all training, inference testing, and integration steps were performed on a single local machine. This constraint required centralized development, and thus all Git operations (including commits and pushes) originated from the same device.

Advantages of this approach:

  • Ensured consistent CUDA and PyTorch environment for model training
  • Avoided serialization/deserialization issues with large-scale checkpoints
  • Direct and quick access to logs, weights, and intermediate outputs

🧠 Model Training Details

  • Object Detection Backbone: Ultralytics YOLOv8
  • OCR Engine: Microsoft TrOCR (Transformer OCR)
  • Auxiliary OCR: Google Cloud Vision API
  • Image Preprocessing & Feature Extraction: Torch + TorchVision

🎯 Model Accuracy

  • mAP@0.5: 0.995 (Mean Average Precision at IoU threshold 0.5)
  • F1 Confidence Curve Score: 1.00
  • Confusion Matrix: Flawless performance on actual Sanskrit manuscript datasets.
    ➕ Refer to the results/ folder for sample predictions and benchmarking screenshots.

🧱 Tech Stack

🔍 AI/ML Frameworks

  • Ultralytics YOLOv8
  • Torch + TorchVision
  • TrOCR
  • Google Cloud Vision API

💻 Frontend

  • TypeScript
  • JavaScript
  • Vite
  • Tailwind CSS

⚙️ Backend

  • Python
  • FastAPI

🧩 Repositories

Layer Repository Link
Frontend GitHub: OBZIUS
Backend GitHub: NeuralSynth
Model Training GitHub: NoiceHax

📁 Results

  • Sample predictions, bounding box overlays, and post-OCR outputs can be found in the results/ directory.

🧪 Future Improvements

  • Migrate training to cloud-based GPUs for higher scalability
  • Add multilingual support and improve layout analysis
  • Build a full user-facing dashboard for batch PDF/document OCR

About

AI-powered OCR and analysis tool for Sanskrit and Indic manuscript digitization and research.

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Used by

Contributors

Languages