Library Occupancy Detection App

Role

ML Engineer / Mobile Dev

Area

Machine Learning

Duration

Jan 2024 – May 2024

Skills

PyTorch · YOLOv5 · React Native · Computer Vision

Overview

Developed as part of a team project, this mobile application uses computer vision and machine learning to help students find available study spaces in university libraries. The app processes real-time camera feeds to detect occupancy levels and provides instant updates to users.

Contributions

Trained and fine-tuned YOLOv5 models for people detection Implemented data preprocessing and augmentation pipeline Built React Native views for occupancy summaries and alerts Integrated on-device inference with efficient batching

Impact

Helped students find study spaces 60% faster Reduced overcrowding in popular library areas Improved overall library experience for students