| 摘要 |
A single still input image is converted into a decomposition video that, when played, appears to be a close facsimile of the input image. Each frame of the decomposition video has a subset of the pixels of the input image that is disjoint from the subset of pixels selected for any other frame. A union of the subsets, represented by each decomposition video frame, contains all the pixels of the input image. To preserve sufficient brightness, a decomposition video generally needs to contain a relatively small number of frames. To achieve effective and efficient blocking, of the content of the input image as it appears in each frame of a decomposition video, the present invention focuses upon a spatial filtering strategy and, preferably, a two-tiered strategy. A first tier focuses upon the obscuring of relatively high frequency spatial frequencies, while a second tier focuses upon the obscuring of relatively low frequency spatial frequencies. |
| 主权项 |
1. A method for processing an input image, comprising:
selecting, performed at least in part with a configuration of computing hardware and programmable memory, vixel subsets of the input image, such that each subset is, at least, substantially disjoint from any other subset selected; selecting, performed at least in part with a configuration of computing hardware and programmable memory, vixel subsets of the input image, such that a union of the subset contains all the vixels of the input image; selecting, performed at least in part with a configuration of computing hardware and programmable memory, a first set of vixel subsets in accordance with subregions of the input image of a first type, that are sized to provide higher-frequency spatial frequency filtering; selecting, performed at least in part with a configuration of computing hardware and programmable memory, two or more subsets, from each member of the first set of vixel subsets, in accordance with subregions of the input image of a second type, that are sized to provide lower-frequency spatial frequency filtering; and playing as a sequence of frames, performed at least in part with a configuration of computing hardware and programmable memory, each of the two or more subsets formed, from each member of the first set of vixel subsets. |
