Deep Image Homography Estimation
Preprint
Abstract
We present a deep convolutional neural network for estimating the relative homography between a pair of images. Our feed-forward network has 10 layers, takes two stacked grayscale images as input, and produces an 8 degree of freedom homography which can be used to map the pixels from the first image to the second. We present two convolutional neural network architectures for HomographyNet: a regression network which directly estimates the real-valued homography parameters, and a classification network which produces a distribution over quantized homographies. We use a 4-point homography parameterization which maps the four corners from one image into the second image. Our networks are trained in an end-to-end fashion using warped MS-COCO images. Our approach works without the need for separate local feature detection and transformation estimation stages. Our deep models are compared to a traditional homography estimator based on ORB features and we highlight the scenarios where HomographyNet outperforms the traditional technique. We also describe a variety of applications powered by deep homography estimation, thus showcasing the flexibility of a deep learning approach.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Exploring Representation Learning With CNNs for Frame-to-Frame Ego-Motion EstimationIEEE Robotics and Automation Letters, 2015
- FlowNet: Learning Optical Flow with Convolutional Networks2015
- ORB-SLAM: A Versatile and Accurate Monocular SLAM SystemIEEE Transactions on Robotics, 2015
- LSD-SLAM: Large-Scale Direct Monocular SLAM2014
- Microsoft COCO: Common Objects in Context2014
- ORB: An efficient alternative to SIFT or SURF2011
- Discovering Higher Level Structure in Visual SLAMIEEE Transactions on Robotics, 2008
- Multiple View Geometry in Computer Vision2004
- A flexible new technique for camera calibrationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Video mosaics for virtual environmentsIEEE Computer Graphics and Applications, 1996