A survey of structure from motion.

<jats:p>The structure from motion (SfM) problem in computer vision is to recover the three-dimensional (3D) structure of a stationary scene from a set of projective measurements, represented as a collection of two-dimensional (2D) images, via estimation of motion of the cameras corresponding to these images. In essence, SfM involves the three main stages of (i) extracting features in images (<jats:italic>e.g.</jats:italic> points of interest, lines,<jats:italic>etc.</jats:italic>) and matching these features between images, (ii) camera motion estimation (<jats:italic>e.g.</jats:italic> using relative pairwise camera positions estimated from the extracted features), and (iii) recovery of the 3D structure using the estimated motion and features (<jats:italic>e.g.</jats:italic> by minimizing the so-called<jats:italic>reprojection error</jats:italic>). This survey mainly focuses on relatively recent developments in the literature pertaining to stages (ii) and (iii). More specifically, after touching upon the early factorization-based techniques for motion and structure estimation, we provide a detailed account of some of the recent camera<jats:italic>location</jats:italic>estimation methods in the literature, followed by discussion of notable techniques for 3D structure recovery. We also cover the basics of the<jats:italic>simultaneous localization and mapping</jats:italic>(SLAM) problem, which can be viewed as a specific case of the SfM problem. Further, our survey includes a review of the fundamentals of feature extraction and matching (<jats:italic>i.e.</jats:italic> stage (i) above), various recent methods for handling ambiguities in 3D scenes, SfM techniques involving relatively uncommon camera models and image features, and popular sources of data and SfM software.</jats:p>