Open Jackandrome opened 5 years ago
I want to use ORNconv to encode orientation informations in object detection. it can be used in a pretrained network?
在实例分割中,想做到对像素级别的方向信息编码,用ORN 是否可行呢?
@Jackandrome 分割不太懂。。有一篇用这个做检测的Rotation-Sensitive Regression for Oriented Scene Text Detection可以参考
I want to use ORNconv to encode orientation informations in object detection. it can be used in a pretrained network?