yolov8-face-landmarks-opencv-dnn

yolov8-face-landmarks-opencv-dnn copied to clipboard

AttributeError: 'tuple' object has no attribute 'flatten'

2

图片无人脸情况下，cv2.dnn.NMSBoxes输出的indices是空的tuple()。 有人脸情况下输出的是array([2])。 所以if len(indices) > 0: 判断这里是无效的，会在上一句直接报错。烦请修改一下

EscaticZheng

关于输出类别的问题

3

我想请教一下如何获得被识别物体的类别，我用cls_id但得到的不是正确的数字，谢谢

ccy2077

fqa model reference？

2

Hi，Could you provide the reference of the fqa model ([face-quality-assessment.onnx](https://github.com/hpc203/yolov8-face-landmarks-opencv-dnn/blob/main/weights/face-quality-assessment.onnx))? I find the output of model has 10 dimensions, is there any introduction? THX~

carter54

如何更改为GPU版本？

2

您好，现有代码是基于CPU的版本，请问如何将其修改为基于GPU的版本？

bowenzhao-zju

我用export 导出yolov8n-face到onnx没问题,但是yolov8-lite-s and yolov8-lite-t就报错，大佬你是咋导出的呀 ```yolo mode=export model=yolov8-lite-s.pt format=onnx simplify=True batch=1``` ``` Traceback (most recent call last): File "/home/pcb/.local/bin/yolo", line 8, in sys.exit(entrypoint()) File "/home/pcb/.local/lib/python3.8/site-packages/ultralytics/yolo/cfg/init.py", line 378, in entrypoint model =...

pcb9382

大佬，你试过tensorrt加速吗，如果要使用tensorrt，应该怎么处理呀

liuhuan1111

我用yolov8-face训练出来的 关键点只有4个的模型 在pytorch上跑的坐标都是正确的 但是在dnn调用后的坐标混乱 也可能是关键点相关代码需要特殊修改？

szufafa

报错，无法运行

4


JHC000abc

Please explain this error: ValueError operands could not be broadcast together with shapes (6400,) (1600,)

2

#### I'm running `main.py` and getting this error. ``` yolov8-face-landmarks-opencv-dnn on main via 🐍 v3.8.10 ❯ py main.py Traceback (most recent call last): File "main.py", line 171, in boxes, scores,...

AbdulSamadMalik

能不能出一期yolov8目标追踪的c++ onnxruntime部署方法呀

huangqi111