Open-source the deep homography code.

PiperOrigin-RevId: 238620744

deep_homography/README.md

@@ -0,0 +1,10 @@
+# Multi-scale Homography Estimation using Deep Neural Network
+
+__Warning: WORK IN PROGRESS. Currently the code is for demonstration purpose and does not run at the moment.__
+
+This project extends the deep homography estimation method from DeTone et al. [1] with a multi-scale strategy. Given a pair of input images, a homography is first estimated at the lowest resolution and then is progressively refined at higher resolutions. The training can be conducted using a synthetic dataset derived from the MS-COCO benchmark or other image/video datasets by following the method from DeTone et al [1].
+
+The code builds upon Tensorflow(https://www.tensorflow.org/).
+
+[1] DeTone, Daniel, Tomasz Malisiewicz, and Andrew Rabinovich. "Deep image homography estimation." arXiv preprint arXiv:1606.03798 (2016).
+https://arxiv.org/abs/1606.03798

deep_homography/__init__.py

@@ -0,0 +1,16 @@
+# coding=utf-8
+# Copyright 2019 The Google Research Authors.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+