The generation of high-resolution mosaics from multiple images is presented.
The estimation of the high-resolution signal utilizes a motion-compensated filtering approach
for video frames that are affected by spatial aliasing. For that, global motion models and the
corresponding estimation algorithms are employed to provide a very accurate and continuous description
of sub-pixel motion for the background of the video signal. The motion information is exploited
to generate mosaics with four or sixteen times the resolution of the constructing video sequence
or known mosaicing techniques. The mosaicing results indicate that the presented multi-frame methods
provide superior visual quality. The approach is further extended to the generation of high-resolution video.
Fig. 1: Process of mosaicing: warping and blending all frames of a video sequence towards
a common reference system, controlled by estimated global motion parameters.
Fig. 2: High-resolution mosaicing: transformation of video pixels into mosaic of double resolution in both directions, without interpolation of intensity values.
Fig. 3: Successive filling of a mosaic of double resolution in both directions (704x480 pixel), for sequence Stefan (SIF, 352x240 pixel).
Results when processing 1, 5, 15, 49 frames are shown.
Download Video: Succesive filling of a high resolution mosaic
Fig. 4: Mosaic (2040x1020 pixel) with 16 times the resolution of sequence Mobile & Calendar (SIF,
360x240 pixel) for all 300 frames (top: high-resolution mosaicing,
bottom: conventional mosaicing and up-sampling)
Fig. 5: Details from images with 16 times the resolution of the original, for frame no. 150 of sequence Mobile & Calendar, left: bilinear interpolation from original image, right: high-resolution mosaicing.
Download Video: Comparison of progressive TV (720/704x480 pixel) sequences generated form SIF (360/352x240 pixel) sequences using high resolution video mosaicing and bilinear interpolation.
Start
Organisation