A Comparative Analysis of Supervised, Semi Supervised and Unsupervised Object Detection Techniques in Video Surveillance

Object detection plays a critical role in various computer vision applications, from autonomous driving to surveillance and medical imaging. This paper presents a comparative analysis of three prominent object detection techniques: supervised, semi-supervised, and unsupervised learning in the field of video surveillance using UCSD Ped1 and UCSD Ped 2 datasets. Supervised learning has been the most widely used approach due to its high precision, though it requires large amounts of labeled data, which can be time-consuming and expensive to obtain. In contrast, semi-supervised learning aims to leverage both labeled and unlabeled data, offering a balance between unsupervised methods and supervised techniques. Finally, unsupervised learning approaches attempt to identify objects without any labeled data, typically by discovering patterns and structures in the data autonomously. This analysis evaluates each technique in terms of mean average precision and processing time, highlighting their strengths and weaknesses. The paper concludes by discussing potential future directions for improving these methods and integrating them for more robust object detection systems.