ReIDetect: Enhancing security with AI surveillance

The increasing need for security and surveillance has led to the development of various technologies for identifying individuals. However, traditional methods such as facial recognition, gait analysis, and iris scanning have limitations that make them ineffective in specific scenarios. In this Project, we propose a web application for person re-identification that will use body shape, clothing, and gait patterns. Our proposed solution aims to overcome the limitations of traditional methods and make the technology more practical for security and surveillance purposes. The objectives of this proposed solution include re-identifying people using web applications, videos, live cameras, and multiple camera views, as well as tracing the paths of re-identified individuals. This proposed solution will make the technology more accessible and user-friendly for everyone.

Project Purpose

The purpose of this project is to develop a web application that utilizes ReID technology for person re-identification. The primary objective is to provide a user-friendly interface that allows individuals to easily identify specific persons from video footage or live camera feeds. Unlike traditional facial recognition systems, which heavily rely on clear facial images, our proposed solution takes into account additional factors such as body width, height, clothing, and mannerisms. By considering these characteristics, the application can identify individuals even when their faces are blurred or not visible to the camera, significantly enhancing the accuracy and effectiveness of the identification process.

In Figure 1 if a person is in front of a camera system can identify a person by using facial Recognition. What if the person's face is not clear to the camera due to the camera resolution being low, or the person wearing a mask, or the person is too far from the camera? or person’ back, left side, or right side to the camera then how system identify the person?

In this video our application find target person from video