Person Re-Identification From Video Surveillance Systems Using Artificial Intelligence Methods

The study explores use of deep learning models in person re identification, leveraging the advancements made in face recognition however the abundance of model choices presents a challenge in selecting the optimal architecture. The study proposes a comprehensive framework for evaluating deep learning models on person re-identification tasks by considering various performance metrics, dataset preprocessing methods, model architectures, and evaluation techniques to enable a systematic comparison of different approaches through empirical analyses on standard person re-identification datasets. The proposed framework is worked-out in uncovering the strengths and limitations of diverse deep learning strategies. The primary objective is to utilize face recognition methodologies to achieve accurate person re-identification.