Multiresolution View Morphing in the Wavelet Domain

This paper presents a new view-synthesis technique using the 2D discrete wavelet-based view morphing. The view morphing is completely based on pairwise images without the camera calibration and depth information of images. First a Fundamental Matrix related to any pair of images is estimated. Then using the fundamental matrix, the pair of image planes is rectified to be parallel with their corresponding points lying on the same scanline, giving an opportunity to generate new views with linear interpolation techniques. The pre-warped images are then decomposed into a hierarchical structure with the wavelet transform. Corresponding coefficients between two decomposed images are linearly interpolated to form the multiresolution representation of an intermediate view. Quantization techniques [10,11] can be embedded here to compress the coefficients for the purpose of reducing the morphing complexity. Finally, during the display, compressed images are decoded and an inverse wavelet transform is applied. A postwarping procedure is employed to transform the interpolated views to its desired position.