Aperture Supervision for Monocular Depth Estimation
Authors:
- Pratul P. SrinivasanPSPratul P. Srinivasan
- Rahul GargRGRahul Garg
- Neal WadhwaNWNeal Wadhwa
- Ren NgRNRen Ng
- Jonathan T. BarronJBJonathan T. Barron
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 6393-6401
Abstract
We present a novel method to train machine learning algorithms to estimate scene depths from a single image, by using the information provided by a camera's aperture as supervision. Prior works use a depth sensor's outputs or images of the same scene from alternate viewpoints as supervision, while our method instead uses images from the same viewpoint taken with a varying camera aperture. To enable learning algorithms to use aperture effects as supervision, we introduce two differentiable aperture rendering functions that use the input image and predicted depths to simulate the depth-of-field effects caused by real camera apertures. We train a monocular depth estimation network end-to-end to predict the scene depths that best explain these finite aperture images as defocus-blurred renderings of the input all-in-focus image.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Deeper Depth Prediction with Fully Convolutional Residual Networks2016
- Deep Stereo: Learning to Predict New Views from the World's Imagery2016
- Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture2015
- Fast bilateral-space stereo for synthetic defocus2015
- Real‐time Depth of Field Rendering via Dynamic Light Field Generation and FilteringComputer Graphics Forum, 2010
- Defocus MagnificationComputer Graphics Forum, 2007
- Shape-from-shading: a surveyIEEE Transactions on Pattern Analysis and Machine Intelligence, 1999
- Light field rendering1996
- Distributed ray tracingACM SIGGRAPH Computer Graphics, 1984
- Recovering surface shape and orientation from textureArtificial Intelligence, 1981