Therefore, we built our dataset from multiple recordings of real-world scenarios to train YOLOv4 and YOLOv4tiny models. We used a dataset consisting of 3.1k ...
確定! 回上一頁