I am new to deep learning and computer vision. I have a problem where I use the YOLO to detect objects.
For my problem, I just want to recognize 1 human only. So, I changed the final YOLO's layer (which contained 80 neurons) to only 1 neuron, and do the training process with transfer learning techniques. Of course, I do not use the final layer's weights, and these weights are randomly initialized for my problem. I feed only the human data to the model.
However, I realize that after longer training, the model becomes worse. It starts to recognize other objects as a human.
Should I also feed non-human data to the model?