This package provides a Matlab implementation of ACCV2016 paper: "Divide and Conquer: Effcient Density-Based Tracking of 3D Sensors in Manhattan Worlds" for the purpose of study.
-
preprocessing
Computing the 3D patches (super point cloud) in advance for efficient experiment.
-
tracking
Core functions of the motion estimation algorithm.
-
evaluation
A script for evaluating the result and outputing figures.
choose a proper directory and clone by:
git clone https://github.com/Ethan-Zhou/MWO.git
-
Download a dataset from https://vision.in.tum.de/data/datasets/rgbd-dataset/download, fr3 cabinet is recommanded.
-
Run preprocessing/Preprocessing.m which gives you 3D patches saved as independent .mat files.
-
Run tracking/main.m which will give you the 3D motion estimation result.
-
You can tune parameters in load_param_MFVO. Have fun!
The approach is descirbed in the following publication:
-
Divide and Conquer: Effcient Density-Based Tracking of 3D Sensors in Manhattan Worlds (Yi Zhou, Laurent Kneip, Cristian Rodriguez, Hongdong Li), The 13th Asian Conference on Computer Vision (ACCV 2016), Oral presentation.
-
Real Time Rotation Estimation for Dense Depth Senors in Piece-wise Planar Environments (Yi Zhou, Laurent Kneip, Hongdong Li)Intelligent Robots and Systems (IROS), 2016 IEEE/RSJ International Conference on.
You can find the paper at http://users.cecs.anu.edu.au/~u5535909/.
The package is licenced under the MIT License, see http://opensource.org/licenses/MIT.