shudali / rgbd_pose_estimation

A header only library with example files to deliver moer accurate, precise and robust pose estimation than using traditional Perspective-n-Point approaches.

rgbd_pose_estimation (Version 1.2 - 28th of Dec 2017)

Update notes:

• Adapt to the latest Sophus library.
• Remove dependency on boost library.
• Fix a few bugs in the testing example.

Release highlights:

• It is a header only library. (No linking required)
• Example codes available.
• More accurate, precise and robust absolute pose estimation than using traditional Perspective-n-Point approaches or Absolute orientation approaches.
• Tested on Windows and Linux.

Build Notes:

Windows:

Requirements:
	1) Microsoft Visual C++
	   From: https://support.microsoft.com/en-us/kb/2977003/
	2) Eigen
	   From: http://eigen.tuxfamily.org/index.php?title=Main_Page
	   
Optional requirements for the sample codes:
	1) OpenCV 3.0 
	   From: http://opencv.org/
    2) CMake and CMake-GuI (To create project files)
	   From: http://www.cmake.org/

Building rgbd_pose_estimation:
    As the library is composed of a bunch of headers of C++, no linking is needed.

Building the example
    1) Install the OpenCV 3.0 Library 
	2) Create a folder 'build' 
	3) Run CMake-Gui and specify the path to dependences. 
	4) Press 'configure' and then 'generate' button
	5) Open the 'RGBDPoseEstimation.sln' and compile it.

Linux:

Requirements:
	1) Eigen
	   From: http://eigen.tuxfamily.org/index.php?title=Main_Page
	   
Optional requirements for the sample codes:
	1) OpenCV 3.0 
	   From: http://opencv.org/
    2) CMake and CMake-GuI (To create project files)
	   From: http://www.cmake.org/

Building rgbd_pose_estimation:
    As the library is composed of a bunch of headers of C++, no linking is needed.

Building the example
    1) Install the OpenCV 3.0 Library 
	2) Create a folder 'build' 
	3) Run CMake-Gui and specify the path to dependences. 
	4) Press 'configure' and then 'generate' button
	5) make

rgbd_pose_estimation's Issues

[Question] Example with real images

Hei,

Are you still maintaining this library? The codes seems pretty neat, but the demo TestMain.cpp generate synthetic data instead of using real images (RGB+Depth), I am wondering, can you give a through example with real images, like images from rgbd-dataset. An potential example is like pose_estimation_3d2d.cpp.

Looking forward to your reply!

missing includes?

Hi there, i am trying to run your library, but am seeing these missing files...

include "Utility.hpp"

include "AbsoluteOrientation.hpp"

include "AbsoluteOrientationNormal.hpp"

include "PnPPoseAdapter.hpp"

include "AOPoseAdapter.hpp"

include "NormalAOPoseAdapter.hpp"

include "P3P.hpp"

include "MinimalSolvers.hpp"

include "Simulator.hpp"

I have linked boost and eigen, what am I missing here?

Thank you!