The subsequent disquisition unveils an innovative tracking stratagem, specifically tailored for automotive monitoring situations. The recommended tactic aspires to perfect the blueprint, augmenting the detection briskness via a judicious synthesis of YOLOv4 and DeepSORT. The conventional convolution integral within YOLOv4 succumbs to depthwise separable convolution, culminating in a diminished computational exertion and accentuated recognition speed. In addition, the CSPNet framework nestled within YOLOv4 undergoes a tweaking process to pare down the parameter quantity of the blueprint. Alterations in the attribute extraction network of the Deep SORT algorithm facilitate effective re-identification, rendering it ideally malleable for automotive pursuit situations. Experimental excursions on custom datasets signify that despite a slight depreciation in average precision, the detection frame rate skyrockets to an impressive 132.4 FPS, thereby catering to tracking task prerequisites on gadgets with finite performance aptitude. As a result, this academic endeavor proffers a convenient resolution for real-time automotive tracking in settings operating under resource restrictions.