2021.2.6 此项目不再更新，新项目地址: Yolo-Fastest： Faster and stronger https://github.com/dog-qiuqiu/Yolo-Fastest

• Yolo-Fastest： Faster and stronger https://github.com/dog-qiuqiu/Yolo-Fastest
• 此表中NCNN基准未更新最新ARM82数据，最新版本NCNN理论上ARM82会有一倍速度提升，待更新...
• 添加基于ncnn的106关键点 C sample:https://github.com/dog-qiuqiu/MobileNet-Yolo/tree/master/sample/ncnn

***Darknet Group convolution is not well supported on some GPUs such as NVIDIA PASCAL!!!

• AlexeyAB/darknet#6091 (comment)

针对某些Pascal显卡例如1080ti在darknet上 训练失败/训练异常缓慢/推理速度异常 的可以采用Pytorch版yolo3框架 训练/推理

• https://github.com/dog-qiuqiu/yolov3

MobileNetV2-YOLOv3-Lite&Nano Darknet

Mobile inference frameworks benchmark (4*ARM_CPU)

Network	VOC mAP(0.5)	COCO mAP(0.5)	Resolution	Inference time (NCNN/Kirin 990)	Inference time (MNN arm82/Kirin 990)	FLOPS	Weight size
MobileNetV2-YOLOv3-Lite(our)	73.26	37.44	320	28.42 ms	18 ms	1.8BFlops	8.0MB
MobileNetV2-YOLOv3-Nano(our)	65.27	30.13	320	10.16 ms	5 ms	0.5BFlops	3.0MB
MobileNetV2-YOLOv3	70.7	&	352	32.15 ms	& ms	2.44BFlops	14.4MB
MobileNet-SSD	72.7	&	300	26.37 ms	& ms	& BFlops	23.1MB
YOLOv5s	&	56.2	416	150.5 ms	& ms	13.2BFlops	28.1MB
YOLOv3-Tiny-Prn	&	33.1	416	36.6 ms	& ms	3.5BFlops	18.8MB
YOLOv4-Tiny	&	40.2	416	44.6 ms	& ms	6.9BFlops	23.1MB
YOLO-Nano	69.1	&	416	& ms	& ms	4.57BFlops	4.0MB

• Support mobile inference frameworks such as NCNN&MNN
• The mnn benchmark only includes the forward inference time
• The ncnn benchmark is the forward inference time + post-processing time(NMS...) of the convolution feature map.
• Darknet Train Configuration: CUDA-version: 10010 (10020), cuDNN: 7.6.4,OpenCV version: 4 GPU:RTX2080ti

MobileNetV2-YOLOv3-Lite-COCO Test results

Application

Ultralight-SimplePose

• A ultra-lightweight human body posture key point prediction model designed for mobile devices, which can cooperate with MobileNetV2-YOLOv3-Nano to complete the human body posture estimation task
• https://github.com/dog-qiuqiu/Ultralight-SimplePose

YoloFace-500k: 500kb yolo-Face-Detection

Network	Resolution	Inference time (NCNN/Kirin 990)	Inference time (MNN arm82/Kirin 990)	FLOPS	Weight size
UltraFace-version-RFB	320x240	&ms	3.36ms	0.1BFlops	1.3MB
UltraFace-version-Slim	320x240	&ms	3.06ms	0.1BFlops	1.2MB
yoloface-500k	320x256	5.5ms	2.4ms	0.1BFlops	0.52MB
yoloface-500k-v2	352x288	4.7ms	&ms	0.1BFlops	0.42MB

• 都500k了，要啥mAP:sunglasses:
• Inference time (DarkNet/i7-6700):13ms
• The mnn benchmark only includes the forward inference time
• The ncnn benchmark is the forward inference time + post-processing time(NMS...) of the convolution feature map.

Wider Face Val

Model	Easy Set	Medium Set	Hard Set
libfacedetection v1（caffe）	0.65	0.5	0.233
libfacedetection v2（caffe）	0.714	0.585	0.306
Retinaface-Mobilenet-0.25 (Mxnet)	0.745	0.553	0.232
version-slim-320	0.77	0.671	0.395
version-RFB-320	0.787	0.698	0.438
yoloface-500k-320	0.728	0.682	0.431
yoloface-500k-352-v2	0.768	0.729	0.490

• yoloface-500k-v2：The SE&CSP module is added
• V2 does not support MNN temporarily
• wider_face_val(ap05): yoloface-500k: 53.75 yoloface-500k-v2: 56.69

YoloFace-500k Test results(thresh 0.7)

YoloFace-500k-v2 Test results(thresh 0.7)

YoloFace-50k: Sub-millisecond face detection model

Network	Resolution	Inference time (NCNN/Kirin 990)	Inference time (MNN arm82/Kirin 990)	Inference time (DarkNet/R3-3100)	FLOPS	Weight size
yoloface-50k	56x56	0.27ms	0.31ms	0.5 ms	0.001BFlops	46kb

• For the close-range face detection model in a specific scene, the recommended detection distance is 1.5m

YoloFace-50k Test results(thresh 0.7)

YoloFace50k-landmark106(Ultra lightweight 106 point face-landmark model)

Network	Resolution	Inference time (NCNN/Kirin 990)	Inference time (MNN arm82/Kirin 990)	Weight size
landmark106	112x112	0.6ms	0.5ms	1.4MB

• Face detection: yoloface-50k Landmark: landmark106

YoloFace50k-landmark106 Test results

Reference&Framework instructions&How to Train

• https://github.com/AlexeyAB/darknet
• You must use a pre-trained model to train your own data set. You can make a pre-trained model based on the weights of COCO training in this project to initialize the network parameters
• 交流qq群:1062122604

About model selection

• MobileNetV2-YOLOv3-SPP: Nvidia Jeston, Intel Movidius, TensorRT，NPU，OPENVINO...High-performance embedded side
• MobileNetV2-YOLOv3-Lite: High Performance ARM-CPU，Qualcomm Adreno GPU， ARM82...High-performance mobile
• MobileNetV2-YOLOv3-NANO： ARM-CPU...Computing resources are limited
• MobileNetV2-YOLOv3-Fastest： ....... Can you do personal face detection???It’s better than nothing

NCNN conversion tutorial

• Benchmark:https://github.com/Tencent/ncnn/tree/master/benchmark
• NCNN supports direct conversion of darknet models
• darknet2ncnn: https://github.com/Tencent/ncnn/tree/master/tools/darknet

NCNN C++ Sample

• https://github.com/dog-qiuqiu/MobileNetv2-YOLOV3/tree/master/sample/ncnn

NCNN Android Sample

• https://github.com/dog-qiuqiu/Android_MobileNetV2-YOLOV3-Nano-NCNN
• APK:https://github.com/dog-qiuqiu/Android_MobileNetV2-YOLOV3-Nano-NCNN/blob/master/app/release/MobileNetv2-yolov3-nano.apk

DarkNet2Caffe tutorial

Environmental requirements

• Python2.7
• python-opencv
• Caffe(add upsample layer https://github.com/dog-qiuqiu/caffe)
• You have to compile cpu version of caffe！！！

    cd darknet2caffe/
    python darknet2caffe.py MobileNetV2-YOLOv3-Nano-voc.cfg MobileNetV2-YOLOv3-Nano-voc.weights MobileNetV2-YOLOv3-Nano-voc.prototxt MobileNetV2-YOLOv3-Nano-voc.caffemodel
    cp MobileNetV2-YOLOv3-Nano-voc.prototxt sample
    cp MobileNetV2-YOLOv3-Nano-voc.caffemodel sample
    cd sample
    python detector.py
  

MNN conversion tutorial

• Benchmark:https://www.yuque.com/mnn/cn/tool_benchmark
• Convert darknet model to caffemodel through darknet2caffe
• Manually replace the upsample layer in prototxt with the interp layer
• Take the modification of MobileNetV2-YOLOv3-Nano-voc.prototxt as an example

	#layer {
	#    bottom: "layer71-route"
	#    top: "layer72-upsample"
	#    name: "layer72-upsample"
	#    type: "Upsample"
	#    upsample_param {
	#        scale: 2
	#    }
	#}
	layer {
	    bottom: "layer71-route"
	    top: "layer72-upsample"
	    name: "layer72-upsample"
	    type: "Interp"
	    interp_param {
		height:20  #upsample h size
		width:20   #upsample w size
	    }
	}

• MNN conversion: https://www.yuque.com/mnn/cn/model_convert

Thanks

• https://github.com/shicai/MobileNet-Caffe
• https://github.com/WZTENG/YOLOv5_NCNN
• https://github.com/AlexeyAB/darknet
• https://github.com/Tencent/ncnn
• https://gluon-cv.mxnet.io/