Input image Compensated image Remove camera

zooming effect Figure 1: Image compensation for removing the camera zooming effect. The left image is an input image. The right image is an com-pensated image by using the estimated focal length change. SLAM algorithms are widely used in augmented reality applica-tions for registering virtual objects. Most SLAM algorithms esti-mate camera poses and 3D positions of feature points using known intrinsic camera parameters that are calibrated and fixed in advance. This assumption means that the algorithm does not allow chang-ing the intrinsic camera parameters during runtime. We propose a method for handling focal length changes in the SLAM algorithm. Our method is designed as a pre-processing step for the SLAM al-gorithm input. In our method, the change of the focal length is esti-mated before the tracking process of the SLAM algorithm. Camera zooming effects in the input camera images are compensated for by using the estimated focal length change. By using our method, cam-era zooming can be used in the existing SLAM algorithms such as PTAM [4] with minor modifications. In the experiment, the effec-tiveness of the proposed method was quantitatively evaluated. The results indicate that the method can successfully deal with abrupt changes of the camera focal length