search

shekkizh / FCN.tensorflow

Tensorflow implementation of Fully Convolutional Networks for Semantic Segmentation (http://fcn.berkeleyvision.org)
MIT License
1.25k stars 527 forks source link

why you read annotation only one channel,Not three channels?? #73

Closed liuzhisheng1226 closed 6 years ago

liuzhisheng1226 commented 6 years ago

self.annotations = np.array( [np.expand_dims(self._transform(filename['annotation']), axis=3) for filename in self.files])

MarkYangjiayi commented 6 years ago

@liuzhisheng1226 Bro I have the same question, wonder if you ever found a solution to this problem...

shekkizh commented 6 years ago

annotations are labels. No need for 3 channels. Closing due to inactivity