-2

I train a single class model with yolo5.The dataset which contains 400 images is split into train,val and test dataset.Here is a notebook of train codetrain code. The train get a well result .However,when i build a real-time object detection application with the exported model,the result is not satisfactory,it can recognize the object as expected.But there are many misjudge in the detection.The detection result: result

In the result,there is only one pepsi,why the timestamp and the other small area are recognized as pepsi.

Can someone tell me the reason of the result?

desertnaut
  • 57,590
  • 26
  • 140
  • 166
BieFeNg27
  • 1
  • 3

1 Answers1

0

Your question lacks a lot of specifics, but my guess is that you have not trained on nearly enough images. 400 is just a toy and most likely resulted in overfitting.

You are going to want to look at a dataset like COCO or ImageNet to get reasonably generalized results.

user2793857
  • 40
  • 1
  • 5