Computer Vision-Based Approach for Energy Efficiency in Commercial Buildings

This paper introduces an innovative computer vision-based method to improve demand-side management for reducing building energy consumption. Unlike traditional approaches, it accurately detects occupancy status using existing CCTV infrastructure. Advanced deep learning and computer vision techniques track and count people, informing a controlling strategy integrated with building systems. This method achieves a significant reduction of approximately 4% in energy consumption.