Year-end Sale % OFF 
Abstract
Structure from Motion (SfM) can produce highly detailed 3D reconstructions, but distinguishing real surface roughness from reconstruction noise and geometric inaccuracies has always been a difficult problem to solve. Existing SfM commercial solutions achieve noise removal by a combination of aggressive global smoothing and the reconstructed texture for smaller details, which is a subpar solution when the results are used for surface inspection. Other noise estimation and removal algorithms do not take advantage of all the additional data connected with SfM. We propose a number of geometrical and statistical metrics for noise assessment, based on both the reconstructed object and the capturing camera setup. We test the correlation of each of the metrics to the presence of noise on reconstructed surfaces and demonstrate that classical supervised learning methods, trained with these metrics can be used to distinguish between noise and roughness with an accuracy above 85%, with an additional 5–6% performance coming from the capturing setup metrics. Our proposed solution can easily be integrated into existing SfM workflows as it does not require more image data or additional sensors. Finally, as part of the testing we create an image dataset for SfM from a number of objects with varying shapes and sizes, which are available online together with ground truth annotations.
Highlights
Structure from Motion (SfM) is widely used for visualization and inspection purposes in the building [1,2,3], manufacturing [4] and energy industries [5], as well as for geology [6,7,8] and cultural preservation [9,10,11]
In this paper we propose noise estimation metrics, which can predict noise risk and be used to distinguish noise caused by sub-optimal SfM reconstructions from the inherent roughness of the reconstructed objects
The problem of detecting noise and geometric disturbances of 3D reconstructed meshes resulting from SfM is a non-trivial one
Summary
Structure from Motion (SfM) is widely used for visualization and inspection purposes in the building [1,2,3], manufacturing [4] and energy industries [5], as well as for geology [6,7,8] and cultural preservation [9,10,11]. Because of the reliance of SfM on 2D image data, it is prone to geometric noise and topological defects, if optimal image capturing conditions are not met (Figure 1). This has prompted a number of benchmarks [12,13,14] on the accuracy and robustness of SfM solutions, as well as on the best possible lighting conditions, camera positions, image density and captured object surface characteristics. The main contribution of this paper is the exploration, development and evaluation of a number of metrics for determining if the underlying 3D reconstructed surface is noisy or rough. For testing the proposed metrics, we have created a image dataset from a number of number
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
