Project DetailsPublic spaces, particularly parks, are a vital and significant component of urban life. Due to population growth and urban development, especially in the wake of the COVID-19 pandemic, optimal planning and utilization of available public spaces have become critical for improving quality of life.
Technological advancements in artificial intelligence and computer vision present an opportunity to optimize and enhance intelligent space planning through in-depth analysis of statistics such as the manner, frequency, type, and duration of public space usage, as well as population distribution and other characteristics.
Our project is part of a series of works aimed at developing tools for intelligent planning through the analysis of security camera footage from parks. In part A, we focused on classifying group size and stability over time from video segments using an algorithm that combines the pre-trained YOLOv8 computer vision network for object detection with density-based clustering using DBSCAN. Our follow-up project focuses on identifying and classifying types of activities with a Neural Network.
Given the limited and challenging data at our disposal, which we will detail later, we achieved results that exceeded our expectations. As we will demonstrate, our algorithm successfully clusters groups of people in space and characterizes their spatial location, size, duration of existence, and other useful attributes, while distinguishing between adjacent groups and filtering noise such as unstable groups and temporary overlapping of people.
In the 2nd Part of the project, after intensive and prolonged tagging and the construction of a general network as proof-of-concept, the faculty's tagging software was shut down, and all our tagged data was lost. Consequently, we have not yet been able to achieve the project's goal, but we have confirmed the feasibility of the idea and the required mode of operation.
