This paper proposes the method and system to invisibly embed the depth data of a target object into the image captured with a camera at the same timing as taking its shot. The proposed system consists of a projector to throw the illumination light wherein the invisible periodical pattern has been contained, and a camera to take a shot of target object. The projector and camera were aligned so that these devices could be positioned in different depths along the z-coordinate. This kind of alignment enables us to set the spatial frequency of the invisible periodical pattern superimposed onto the image of target object taken with the camera, changeable depending on the depth of the target object. This kind of spatial frequency determined depending on the depth of the object allow us to estimate the depth of target object taken with the camera, based on the spatial frequency of the periodical pattern being taken with the camera at the same time. Based on the fundamental study described above, since this paper allows us to estimate the spatial frequency of the periodical pattern contained in the image being taken with the camera, we are enabled to simultaneously generate the depth map of the 2D image being captured. A 3D image could thus be constituted with the captured 2D image combined with the depth map generated. The proposed new techniques, employing a traditional -D image format without any modification, are advantageous in easiness of transferring and storing the 3D images over the conventional 3D image generations.
© 2014 IEEEPDF Article