Projects
Ongoing
ArtEdge: Real-Time Neural Style Transfer for Mobile Devices
A mobile-first iOS application demonstrating the feasibility of performing Neural Style Transfer (NST) directly on-device using edge computing principles.
- Developing real-time neural style transfer on hardware-constrained devices
- Achieved near real-time inference speeds (millisecond-level for some models) using hardware acceleration.
Python
Swift
CoreML
Completed
Beacon Of Hope: A Personalized Meal Plan Recommender
A personalized meal plan recommender system that leverages machine learning to provide tailored nutrition recommendations while providing insightful visualizations.
- Developed interactive meal exploration and categorization
- Designed visual comparison and tracking of nutritional information
React.js
Python
Completed
RoostAI: A University Centered Chatbot
A RAG-based chatbot for the University of South Carolina, designed to assist students, faculty, and staff with information and resources.
- Developed a conversational AI system using the RAG framework, enabling natural language interactions with the chatbot
- Designed a user-friendly interface for the chatbot, allowing users to ask questions, get answers, and explore campus services
Python
LLMs + RAG
Completed
Segify: Semantic Segmentation for Localized Artistic Effects
A novel approach combining semantic segmentation with neural style transfer for precise artistic control.
- Developed a novel approach for segment-based neural style transfer, combining AdaIN layers for real-time style transfer with the Segment Anything model for accurate segmentation
- Implemented an interactive user interface, enabling user-guided selection of content, style, target regions, and style loss weights for creative exploration
- The project offered a user-centric solution for artistic image manipulation, surpassing the limitations of traditional methods by providing localized style control
Python
PyTorch
Computer Vision
Neural Networks
Style Transfer
Completed
Deep Learning & Autoencoders for Colorization
An advanced image colorization system using convolutional autoencoders to transform black and white photographs into vibrant color images.
- Developed and trained a convolutional autoencoder for accurate image colorization of black and white photographs using Python and PyTorch
- Trained the model on nearly 28,000 images real-world images from the Google Landmarks dataset
Python
PyTorch
Deep Learning
Computer Vision
Autoencoders