MobileNetV2-YOLOv3-Nano的Darknet实现:移动终端设计的目标检测网络,计算量0.5BFlops!华为P40:MNN_ARM82单次推理时间6ms 模型大小:3MB!yoloface-500k:只有500kb的实时人脸检测模型
yoloface-500k-v2352x2884.7ms&ms0.1BFlops0.42MB 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. ...
MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire: - maoweinuaa/MobileNetv2-YOLOV3
直接从现货分销商和其他供应商中查询MS-500K-B。使用netCOMPONENTS,世界上最大的电子元件查询网站找到MS-500K-B元件和数据表。
https://github.com/Tencent/ncnn https://gluon-cv.mxnet.io/ 简介 MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops500KB:fire::fire::fire: 暂无标签 保存更改 发行版 暂无发行版 贡献者(2) 全部 近期动态 4年多前创建了仓库...
MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire: - FuXiangGit/MobileNet-Yolo
MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire: Topics computer-vision deep-learning cv cnn yolo face-detection object-detection landmark-detection darknet landmark mnn ncnn mobilenetv2 yolov3 mobilenet-yolo ncnn-model mnn-frame...
MobileNetV2-YoloV3-Nano: 0.5BFlops 3MB HUAWEI P40: 6ms/img, YoloFace-500k:0.1Bflops 420KB:fire::fire::fire: - LUOBO123LUOBO123/MobileNet-Yolo
MobileNetV2-YOLOv3-Nano的Darknet实现:移动终端设计的目标检测网络,计算量0.5BFlops!华为P40:MNN_ARM82单次推理时间6ms 模型大小:3MB!yoloface-500k:只有500kb的实时人脸检测模型
yoloface-500k-v2352x2884.7ms&ms0.1BFlops0.42MB 都500k了,要啥mAP😎 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. ...