SSH context module
ilyanelken opened this issue · comments
Hi, I have noticed that in SSH context module all kernels are 3x3 rather than 3x3/5x5/7x7.
Is there a special reason for that?
RetinaFace (Single-stage Dense Face Localisation in the Wild, 2019) implemented (ResNet50, MobileNetV2 trained on single GPU) in Tensorflow 2.0+. This is an unofficial implementation. With Colab.
ilyanelken opened this issue · comments
Hi, I have noticed that in SSH context module all kernels are 3x3 rather than 3x3/5x5/7x7.
Is there a special reason for that?