How to understand the position-sensitive score map? What's the meaning of the translation-invariance in the classification stage and the translation-variance in the object detection stage? How to show it in the object detection detection framework in R-FCN? Thank you very much!
How to understand the position-sensitive score map? What's the meaning of the translation-invariance in the classification stage and the translation-variance in the object detection stage? How to show it in the object detection detection framework in R-FCN? Thank you very much!