fyu / dilation

Dilated Convolution for Semantic Image Segmentation

Home Page:https://www.vis.xyz/pub/dilation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

about the scale

YanhuaCheng opened this issue · comments

As your paper, input 512x512 images, the output is 64x64. But which is not according the train.prototxt. So for the training, the input is also 900x900? Thanks.

The current released model is for testing, normally called deploy.prototxt, instead of training. The training network will be also released. Please stay tuned.