yolov5-face

yolov5-face copied to clipboard

Confusion_matrix map is always wrong when I test with multi-labels. I found the pred is not original coordinates. I think it should be changed like this, and it works well...

mrxRunwuxiwusheng

Yolov5-face is not working via torch hub.

4

Here is an example code: ``` #!/bin/python import torch # Model model = torch.hub.load('deepcam-cn/yolov5-face', 'yolov5s', pretrained=True) # Images imgs = ['https://ultralytics.com/images/zidane.jpg'] # batch of images # Inference results = model(imgs)...

ioctl-user

martinenkoEduard

martinenkoEduard

Could you please share an inference example?

martinenkoEduard

如何测FDDB数据集上检测yolov5face的性能？

2

在论文的F：YOLOv5Face on FDDB Dataset部分中，看到YOLOv5s在FDDB数据集中取得了0.9843的性能，这是如何检测出来的？ 祝您生活愉快，万事顺心。

SmallHedgeHogG

xailxail

There was an error: "NameError: name 'warnings' is not defined". The solution was to simply import the module. Reproduce steps: 1. python3 export.py --weights ./weights/yolov5s.pt --img 640 --batch 1 ```...

DragonflyRobotics

Yolov5-face inference ’camera input’ FPS is lower than normal yolov5?

2

English is not my native language; please excuse typing errors. “我在使用摄像头输入的数据用以推理的时候发现帧数只有5FPS左右，而使用原版的YOLOV5却有20FPS” "When I used the data input from the camera for inference, I found that the frame rate was only...

NoNounknow

hi，看了您的项目觉得很厉害，想请教个类似问题，关于小目标的检测问题，是否和人脸检测有相似性，看了您的论文中有关于对SPPF的kernel size的修改来提升小目标，除此以外对数据集的处理有什么方式呢？比如实际要检测的目标在图片中的像素**占比**为整张图片的1%，是否需要在数据增强中，把训练集的目标也缩放到图片的**同等比例**，这样训练就可以增强对类似尺度的目标的识别**性能**呢？因为硬件资源有限无法完整验证这个问题，希望您提供下思路和经验，非常感谢！

wxkkk