Hi, @lewisluk , thank you for your work. But I have a question about the implementation of Feature Fusion Module(FFM) in model.py, which you directly reduce the input feature to only one channel, and then learn a channel-wise attention. The attention is a scaler as the number of channel is one, that is abnormal. Can you explain it? Thanks a lot.

Would you please tell me where to find the checkpoint or ckpt file？Or is should i obtain it through extra steps?

Anyway, thank you all the same for your code !

Hello,Thank you for sharing your code.I want to know some detail about the loss function.
In the function,what does logits and factor_lambda mean? Is boundary_target the predicted result?
Thanks.
`def refine_loss(logits, labels, boundary_target):
gamma1 = 0.5
gamma2 = 1-gamma1
factor_lambda= 1.5

dy_logits, dx_logits = tf.image.image_gradients(logits)
dy_labels, dx_labels = tf.image.image_gradients(labels)

# magnitudes of logits and labels gradients
Mpred = tf.sqrt(tf.square(dy_logits)+tf.square(dx_logits))
Mimg = tf.sqrt(tf.square(dy_labels)+tf.square(dx_labels))

# define cos loss and mag loss
cosL = (1-tf.abs(dx_labels*dx_logits+dy_labels*dy_logits))*Mpred
magL = tf.maximum(factor_lambda*Mimg-Mpred,0)

# define mask
M_bound = boundary_target/255.

# define total refine loss
refineLoss = (gamma1*cosL + gamma2*magL)*M_bound
return tf.reduce_mean(refineLoss)
# return  tf.reduce_mean(refineLoss), Mpred, Mimg`