I'm a Computer Science Engineering undergraduate student with a strong passion for Artificial Intelligence and especially Computer Vision. I'm always excited to learn more about how computers can "see" and interpret the world.
- ๐ญ Iโm currently diving deep into Semantic Segmentation to understand how to classify each pixel in an image.
- ๐ฑ Iโm currently learning more about advanced topics in Deep Learning and Computer Vision.
- ๐ฏ Iโm looking to collaborate on projects related to AI, Computer Vision, or any interesting software engineering challenges.
- ๐ค Iโm looking for help with understanding advanced concepts in machine learning and finding good research papers.
- ๐ซ How to reach me: [email protected]
Here are a few technologies I've been working with recently:
- Languages: Python, C++, Kotlin, JavaScript
- Frameworks & Libraries: PyTorch, TensorFlow, OpenCV, Scikit-learn
- Tools: Git, Jupyter Notebook
This project is an on-device, AI-powered mobile application designed to help users efficiently manage and organize their smartphone photo libraries. It automatically filters unnecessary photos, clusters similar images, and allows users to create personalized albums based on their own image examples, all while ensuring user privacy by processing everything directly on the device.
-
Key Features
-
Unified Classification: Automatically categorizes photos into classes like 'Blurry', 'Screenshots', 'Documents', etc., for easy filtering.
-
Similar Clustering: Groups together visually similar photos, such as burst shots or duplicates.
-
Custom Albums: Users can train the app with a few sample images to create their own personalized photo albums using Few-Shot Learning.
-
Privacy-Focused: All AI processing is done 100% on-device, meaning your photos never leave your phone.
-
-
Tech Stack
-
Frontend: React Native (with Expo Bare Workflow)
-
Language: Python, TypeScript
-
On-Device AI: TensorFlow Lite
-
Core AI Model: MobileNet (using Transfer Learning)
-
UI Framework: React Native Paper, NativeWind (Tailwind CSS)
-