抄録
This paper proposes a new object detection model, which is based on YOLOX-Nano, but some layers not only inbackbone but also in neck and head parts of YOLOX-Nano are replaced by layers in MobileNetV3. This paper alsoshows the comparison between the proposed model and the original YOLOX-Nano and the other YOLO models interms of number of parameters, inference time, and object detection accuracy
https://ror.org/02f0psx94 
