泡泡一分鐘：Real-Time Vehicle Detection from Short-Range Aerial Image with Compressed MobileNet

張寧 Real-Time Vehicle Detection from Short-Range Aerial Image with Compressed MobileNet
連接：https://pan.baidu.com/s/1Uk6McGCnCzTieJYehjSFBA
提取碼：e0bn

Yuhang He, Ziyu Pan, Lingxi Li, Yunxiao Shan, Dongpu Cao and Long Chen

 Vehicle detection from short-range aerial image faces challenges including vehicle blocking, irrelevant object interference, motion blurring, color variationetc., leading to the difficulty to achieve high detection accuracy and realtime detection speed. In this paper, benefiting from the recent development in MobileNet family network engineering, we propose a compressed MobileNet which is not only internally resistant to the above listed challenges but also gains the best detection accuracy/speed tradeoff when comparing with the original MobileNet. In a nutshell, we reduce the bottleneck architecture number during the feature map downsampling stage but add more bottlenecks during the feature map plateau stage, neither extra FLOPs nor parameters are thus involved but reduced inference time and better accuracy are expected. We conduct experiment on our collected 5-k short-range aerial images,containingsixvehiclecategories:truck,car,bus,bicycle, motorcycle, crowded bicycles and crowded motorcycles. Our proposed compressed MobileNet achieves 110 FPS(GPU), 31 FPS(CPU) and 15 FPS(mobile phone), 1.2 times faster and 2% more accurate(mAP) than the original MobileNet.