Constructing Locally Dense Point Clouds Using OpenSfM and ORB-SLAM2

TL;DR

This paper presents a method to align point clouds generated by ORB-SLAM2 and OpenSfM using textured tags and feature matching, enabling accurate registration of different 3D reconstructions.

Contribution

It introduces a novel approach for registering point clouds from different sources by using textured tags and feature matching to compute the transformation matrix.

Findings

Successful alignment of point clouds from ORB-SLAM2 and OpenSfM

Effective use of textured tags for feature extraction and matching

Improved accuracy in multi-source 3D reconstruction

Abstract

This paper aims at finding a method to register two different point clouds constructed by ORB-SLAM2 and OpenSfM. To do this, we post some tags with unique textures in the scene and take videos and photos of that area. Then we take short videos of only the tags to extract their features. By matching the ORB feature of the tags with their corresponding features in the scene, it is then possible to localize the position of these tags both in point clouds constructed by ORB-SLAM2 and OpenSfM. Thus, the best transformation matrix between two point clouds can be calculated, and the two point clouds can be aligned.