Eigen-unsupported  3.4.90 (git rev a4098ac676528a83cfb73d4d26ce1b42ec05f47c)
TensorImagePatch Class Reference

Detailed Description

Patch extraction specialized for image processing. This assumes that the input has a least 3 dimensions ordered as follow: 1st dimension: channels (of size d) 2nd dimension: rows (of size r) 3rd dimension: columns (of size c) There can be additional dimensions such as time (for video) or batch (for bulk processing after the first 3. Calling the image patch code with patch_rows and patch_cols is equivalent to calling the regular patch extraction code with parameters d, patch_rows, patch_cols, and 1 for all the additional dimensions.

The documentation for this class was generated from the following file: