Scene-Specialized Multitarget Detector with an SMC-PHD Filter and a YOLO Network

<div>Overall framework of the proposed method. The framework input is a generic, fine-tuned YOLO detector. A visual sequence is provided to the scheme without manual labelling. To customize the YOLO network, an iterative method automatically estimates both parameters.</div>

Computational Intelligence and Neuroscience

fig2

Figure 2

Figure 2: Scene-Specialized Multitarget Detector with an SMC-PHD Filter and a YOLO Network 