Reputation: 21
I'm trying to train a model with Yolo v5 to detect multiple objects on sales flyers. each image in the dataset used in training contains only one object and obviously a single bounding box. I'm wondering if that will affect the performance of the model in a bad way? because what I'm trying to do at the end is detecting multiple objects on each sale flyer. Thank you for your help.
Upvotes: 1
Views: 958
Reputation: 1359
It probably will lower your AP if you work like this, but give it a try. It really depends on your training images and your data augmentations. I dont know about YOLOv5 but YOLOv4 has its Mosaic Data Augmentation which will adress your problem in a way i guess.
Upvotes: 1