IMAGE DEBLUR FOR 3D SENSING MOBILE DEVICES

Advanced mobile devices with an application of Google Project Tango which can trace the motion and the orientation of the device and reconstruct the 3D scenes in real time. Since users always use such devices to scan the real 3D scenes under severe camera motion, the image blur leads to the distortion in the depth images of the scene. However, traditional image deblur methods cannot effectively address this problem since they deblur images without considering 3D information in the depth images. In this paper, we use Google Tango mobile devices equipped with 3D sensors to deblur the depth images for the high accuracy of 3D reconstruction in the real 3D scenes. These sensors can provide the precise geometry data of camera motion trajectory such that this allows us to fast recover the unknown parameters of the 3D information on the depth images without huge time-consumption estimations. Experiments show that our algorithm outperforms previous approaches in not only smaller running time but also the better quality of the restored depth image.