Recovery of dense geometry and camera motion from a set of monocular images is a well-known problem that can be solved quite reliably in well-conditioned environments. However, common algorithms assume static lighting conditions and presence of sufficient scene texture, allowing reliable detection and matching of image features. Sometimes, however, these prerequisites are not met, and feature-based fail. We suggest to address the problem by applying a purely intensity-based approach that can be extended to take into account changing lighting conditions. To this end, we have investigated the applicability of sliding-window intensity-based bundle-adjustment methods to this problem.
«
Recovery of dense geometry and camera motion from a set of monocular images is a well-known problem that can be solved quite reliably in well-conditioned environments. However, common algorithms assume static lighting conditions and presence of sufficient scene texture, allowing reliable detection and matching of image features. Sometimes, however, these prerequisites are not met, and feature-based fail. We suggest to address the problem by applying a purely intensity-based approach that can be...
»