Affinity-based distributed algorithm for 3D reconstruction in large scale Visual Sensor Networks

Abstract

In recent years, Visual Sensor Networks have emerged as an interesting category of distributed sensor-actor systems to retrieve data from the observed scene and produce information. Indeed, the request for accurate 3D scene reconstruction in several applications is leading to the development of very large systems and more specifically to large scale motion capture systems. When dealing with such huge amount of data from a large number of cameras it becomes very hard to make real time reconstruction on a single machine.Within this context, a distributed approach for reconstruction on large scale camera networks is proposed. The approach is based on geometric triangulation performed in a distributed fashion on the computational grid formed by the camera network organized into a tree structure. Since the computational performance of the algorithm strongly depends on the order in which cameras are paired, to optimize the reconstruction a pairing strategy is designed that relies on an affinity score among cameras. This score is computed from a probabilistic perspective by studying the variance of the 3D target reconstruction error and resorting to a normalized cut graph partitioning.The scaling laws and the results obtained in simulation suggest that the proposed strategy allows to obtain a significant reduction of the computational time.

An ISVD-based Euclidian structure from motion for smartphones

“…However, as the number of images increases such procedure becomes computationally demanding. Several techniques based on different similarity measures have been previously proposed in the literature to match features only for "highly correlated" images (Furukawa et al, 2010, Goesele et al, 2007, Masiero and Cenedese, 2013. The rationale among such image selection methods is that images to be compared should be reasonably similar to have a large number of common features, however they should also have different point of views (e.g.…”

“…a large baseline) to ensure a good reconstruction of the feature spatial positions. In this paper we take advantage of the information on the device position and orientation to select the small set of images to compare with the current one for feature matching: the selection is done using considerations on the expected reconstruction error modeled similarly to (Masiero and Cenedese, 2013). …”