An overview of small target detection and tracking methods

Object detection and tracking are crucial in computer vision and visual surveillance, enabling the detection, identification, and subsequent tracking of objects within images or video sequences. These tasks underpin surveillance systems, facilitating automated video annotation, identifying important events, and detecting unusual activities. However, the detection and tracking of small objects poses significant challenges in computer vision due to their delicate appearance and limited distinctive features, resulting in a lack of critical information. This deficit complicates the tracking process and often results in reduced efficiency and accuracy. To shed light on the complexities of small object detection and tracking, we comprehensively review existing methods in this field and categorize them from different perspectives. We also provide an overview of existing datasets specifically designed for small object detection and tracking, with the aim of informing and benefiting future research in this area. We outline the most commonly used evaluation metrics to evaluate the performance of small object detection and tracking techniques. Finally, we reviewed the challenges in this field and discussed future trends. By addressing these issues and leveraging future trends, we aim to push the boundaries of small object detection and tracking, thereby enhancing the performance of surveillance systems and expanding their real-world applications.

An overview of small target detection and tracking methods Keywords:

small objects detection , Tracking , Small targets