This repository is a combination of several depth from defocus projects based on OpenCV and the dlib library.
The code in this repository has the following dependecies:
- CMake 2.8.12+
- davemers0160 common code repository
- davemers0160 dlib-contrib repository
- dlib library
- OpenCV v4+
This folder contains the project code that is shared between all of the projects.
This folder contains the project code that runs the performance analysis of a given network against a given dataset.
This folder contains the project code that runs the deep learning training of the DfD-Net network.
This folder contains the project code that applies the graph cuts method to the depth from defocus challenge.
When using this work please cite the following:
3-D SCENE RECONSTRUCTION FOR PASSIVE RANGING USING DEPTH FROM DEFOCUS AND DEEP LEARNING
DataCite
Emerson, David Ross (2019): 3-D SCENE RECONSTRUCTION FOR PASSIVE RANGING USING DEPTH FROM DEFOCUS AND DEEP LEARNING. figshare. Thesis. https://doi.org/10.25394/PGS.8938376.v1
BiBtex
@article{Emerson2019,
author = "David Ross Emerson",
title = "{3-D SCENE RECONSTRUCTION FOR PASSIVE RANGING USING DEPTH FROM DEFOCUS AND DEEP LEARNING}",
year = "2019",
month = "10",
url = "https://hammer.figshare.com/articles/3-D_SCENE_RECONSTRUCTION_FOR_PASSIVE_RANGING_USING_DEPTH_FROM_DEFOCUS_AND_DEEP_LEARNING/8938376",
doi = "10.25394/PGS.8938376.v1"
}