Sampling-Based Multiview Reconstruction without Correspondences for 3D Edges

Abstract

This paper introduces a novel method for featurebased 3D reconstruction using multiple calibrated 2D views. We use a probabilistic formulation of the problem in the 3D, reconstructed space that allows using features that cannot be matched one-to-one, or which cannot be precisely located, such as points along edges. The reconstructed scene, modelled as a probability distribution in the 3D space, is defined as the intersection of all reconstructions compatible with each available view. We introduce a method based on importance sampling to retrieve individual samples from that distribution, as well as an iterative method to identify contiguous regions of high density. This allows the reconstruction of continuous 3D curves compatible with all the given input views, without establishing specific correspondences and without relying on connectivity in the input images, while accounting for uncertainty in the input observations, due e.g. to noisy images and poorly calibrated cameras. The technical formulation is attractive in its flexibility and genericity. The implemented system, evaluated on several very different publicly-available datasets, shows results competitive with existing methods, effectively dealing with arbitrary numbers of views, wide baselines and imprecise camera calibrations.

Wire Structure Image-Based 3d Reconstruction Aided by Deep Learning

“…Some previous works obtained a 3D reconstruction of wire structures in the form of individual curve segments (Teney, Piater, 2012, Usumezbas et al, 2016. In case a dense point cloud can be produced, either with laser scanning or MVS methods, Huang et al (Huang et al, 2013) presented an automated solution to extract curve skeletons based on the L1-medial axis.…”

Image-based reconstruction of wire art

“…Curve-based reconstruction methods can be grouped according to the simplifying assumptions they make: Berthilsson et al (2001) assume that each 3D curve is fully visible in each view with no selfocclusions; Hong et al (2004) focus on reflective symmetric curves that are fully visible in one or two images; assuming access to the range of depth values in the reconstructed scene, Kahl and August (2003) assign depth values to 2D curves, while Teney and Piater (2012) design a probability distribution to model the space of plausible curve reconstructions. Rao et al (2012) explore known correspondences between endpoints of planar curves to develop a curve-based SLAM (simultaneous localization and mapping) system.…”

Power Transmission Line Reconstruction From Sequential Oblique Uav Images

“…The researchers at the Australian Research Centre for Aerospace Automation (ARCAA) have been working on the generation of 3D Occupancy Map of the environment using stereo vision and laser scanning (Hrabar et al, 2010). Although there are many related work in recovering 3D structures form 2D images (Fabbri and Kimia, 2010;Teney and Piater, 2012;Hofer et al, 2015;Hofer et al, 2017;Li et al, 2018;Langlois et al, 2019;Tian et al, 2019), the wire-like thin object reconstruction remain unsolved. The power lines are typical thin and linear structure.…”