Towards BAFOS: Bundle adjustment with feature orientation and scale

This paper proposes a new approach for enhanced-accuracy visionbased pose estimation in GPS-denied unknown environments. The problem is typically addressed using bundle adjustment (BA) approaches, that minimize the difference between the projected and measured image features (re-projection error). We propose to incorporate within bundle adjustment new type of constraints that relate between image feature scale and orientation between matched views. While feature scale and orientation play an important role in image matching, they have not been utilized thus far for estimation purposes in BA framework. Our approach exploits feature scale and orientation information, that is already available from the image matching process, and uses it to enhance the accuracy of bundle adjustment. In this paper we make the first steps in this direction and develop the corresponding feature scale constraint, which relates between image measured scale and the actual scale of the observed landmark. We demonstrate the proposed concept in a statistical simulation.