I'm doing custom object detection and I have one item that is always identified wrong. In short, it's detecting what it thinks is a custom object but isnt. Is the only solution adding more images? Or is there anything else I can do to make the model predictions better.
What's the size of the dataset, how many objects are in the dataset and how varying are the images (e.g different scenes, objects color, lighting conditions)?
I'm doing custom object detection and I have one item that is always identified wrong. In short, it's detecting what it thinks is a custom object but isnt. Is the only solution adding more images? Or is there anything else I can do to make the model predictions better.
Thanks in advance.