Robust Joint Selection of Camera Orientations and Feature Projections over Multiple Views

Abstract

A number of critical factors arises when a complex 3D scene is to be reconstructed by means of a large sequence of different views. Some of them are related to the ability to recover the correct identity and the accurate projection of each observed feature. Other sources of error are tied to the reliability of the orientation estimate for each view. Here we propose a method that tries to solve both problems at the same time, while being also inherently resilient to outliers. At the core of the approach stands a widely adopted game-theoretical selection technique, which has already been successfully adopted to address similar tasks. The original inception, however, has been further refined to allow both better accuracy and tighter selection, as well as a greatly reduced memory consumption and computation complexity. By exploiting these enhancements, we were able to apply our technique to real-wolrd scenarios involving several hundreds of view points and tens of thousands of independent observations.