Skip to content

DispNet correlation layer tensorflow implementation. the code has been tested success!

License

Notifications You must be signed in to change notification settings

luoru/correlation1d-tf

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Correlation1d Tensorflow version

DispNet correlation1d layer tensorflow implementation. the code has been tested success! This code mainly based on (https://github.com/sampepose/flownet2-tf.git), which has implement the correlation of flownet. Other version of correlation1d in github still have some bugs

Compiling

just run compille.sh

correlation1d example

$ python
>>> import tensorflow as tf
>>> x1 = tf.random_normal([1, 10, 10, 4])
>>> x2 = tf.random_normal([1, 10, 10, 4])
>>> conv_x1 = tf.layers.conv2d(x1, 10, 3, 1, 'same', name='conv_x1')
>>> conv_x2 = tf.layers.conv2d(x2, 10, 3, 1, 'same', name='conv_x2')
>>> corr = correlation1d(conv_x1, conv_x1, 1, 4, 1, 1, 4)
>>> grad_corr2x1 = tf.gradients(corr, x1)

About

DispNet correlation layer tensorflow implementation. the code has been tested success!

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published