Module: tf.contrib.image

Defined in tensorflow/contrib/image/__init__.py.

Ops for image manipulation.

API

This module provides functions for image manipulation; currently, chrominance transforms (including changing saturation and hue) in YIQ space and projective transforms (including rotation) are supported.

Image Transformation Ops

Image Segmentation Ops

Matching Ops

Random Dot Stereogram Ops

Functions

angles_to_projective_transforms(...): Returns projective transform(s) for the given angle(s).

bipartite_match(...): Find bipartite matching based on a given distance matrix.

compose_transforms(...): Composes the transforms tensors.

connected_components(...): Labels the connected components in a batch of images.

dense_image_warp(...): Image warping using per-pixel flow vectors.

flat_transforms_to_matrices(...): Converts tf.contrib.image projective transforms to affine matrices.

interpolate_spline(...): Interpolate signal using polyharmonic interpolation.

matrices_to_flat_transforms(...): Converts affine matrices to tf.contrib.image projective transforms.

rotate(...): Rotate image(s) counterclockwise by the passed angle(s) in radians.

single_image_random_dot_stereograms(...): Output a RandomDotStereogram Tensor for export via encode_PNG/JPG OP.

sparse_image_warp(...): Image warping using correspondences between sparse control points.

transform(...): Applies the given transform(s) to the image(s).

translate(...): Translate image(s) by the passed vectors(s).

translations_to_projective_transforms(...): Returns projective transform(s) for the given translation(s).