You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I have some questions about the hyper parameter: pos_num. In paper, you said "2)The projection of this ground-truth object’s center point locate within this feature map bin." But in practice, a torch.topk() function is used to get the positive target location, which is assigned by pos_num=1. And all the configs' pos_num are set to 1. Here comes my questions:
1)Is this configuration similar with that written in your paper?
2)Why only one point is assigned?
3)Did you do ablation experiments about the selection of it?
Thanks for answering!
The text was updated successfully, but these errors were encountered:
Hi, I have some questions about the hyper parameter: pos_num. In paper, you said "2)The projection of this ground-truth object’s center point locate within this feature map bin." But in practice, a torch.topk() function is used to get the positive target location, which is assigned by pos_num=1. And all the configs' pos_num are set to 1. Here comes my questions:
1)Is this configuration similar with that written in your paper?
2)Why only one point is assigned?
3)Did you do ablation experiments about the selection of it?
Thanks for answering!
The text was updated successfully, but these errors were encountered: