A Moving Object Tracking Technique Using Few Frames with Feature Map Extraction and Feature Fusion

Abstract

Moving object tracking techniques using machine and deep learning require large datasets for neural model training. New strategies need to be invented that utilize smaller data training sizes to realize the impact of large-sized datasets. However, current research does not balance the training data size and neural parameters, which creates the problem of inadequacy of the information provided by the low visual data content for parameter optimization. To enhance the performance of moving object tracking that appears in only a few frames, this research proposes a deep learning model using an abundant encoder–decoder (a high-resolution transformer (HRT) encoder–decoder). An HRT encoder–decoder employs feature map extraction that focuses on high resolution feature maps that are more representative of the moving object. In addition, we employ the proposed HRT encoder–decoder for feature map extraction and fusion to reimburse the few frames that have the visual information. Our extensive experiments on the Pascal DOC19 and MS-DS17 datasets have implied that the HRT encoder–decoder abundant model outperforms those of previous studies involving few frames that include moving objects.

Tracking moving targets of interest-based deep learning

“…For example, Lim et al proposes to use color feature points to adjust the size of the search interested region window as a way to solve the problem of similarity between target color and background 8 . Abdulaziz et al proposes to complete the extraction of feature information in a deep learning model using a high-resolution converter 9 . Haris et al proposes to detect the object of interest by means of a filter with maximum average correlation height, while the gradient descent technique is used to optimize the particle filter as a way to help the model converge quickly and improve the accuracy of target tracking 10 .…”