search

rishizek / tensorflow-deeplab-v3-plus

DeepLabv3+ built in TensorFlow
MIT License
833 stars 307 forks source link

the output image size of inference.py isn't the same as the input image's #41

Open bonbonjour opened 5 years ago

bonbonjour commented 5 years ago

I run inference.py to see the output prediction of pascal voc dataset and my own data, and I noticed that the output image size is not the same as the input image, can you please tell me why. thanks! Additionally, if i want to train my own data with 16 class, do I just need to change number class and label_class?? hope to get your reply!

swarmt commented 5 years ago

Have a look at the end of the main function in inference.py

Change plt.axis('off') plt.imshow(mask) plt.savefig(path_to_output, bbox_inches='tight')

To mask.save(path_to_output)

As for training with your own dataset, Search the entire project for _NUM_CLASSES and modify it.