Abstract: A method of animating a sprite in a video production comprising a plurality of sequential video frames is disclosed. The method comprises the steps of selecting (2004) a feature, with which the sprite is to be juxtaposed, in one of said video frames, applying (2006) a feature tracking process to the video production to thereby output (2008), for a series of said plurality of video frames containing the feature, a temporal-spatial record for the feature across the plurality of video frames, and compositing (2010), with the series of said plurality of video frames, a corresponding series of instantiations of the sprite dependent upon the temporal-spatial record.

Abstract: A method (500, 600) of tracking a feature (201) across a sequence of image frames is disclosed. A current feature position in a current frame is estimated from at least a previous feature position in a previous frame. Feature data is extracted from pixel data of the current frame. The feature data is then compared with reference data. If a difference between the feature data and the reference data is larger than a predetermined number, then track of the feature (201) has been lost. Alternatively, the current feature position is appended to the trajectory of the feature, and the reference data is updated periodically with feature data of a plurality of frames. Preferably the reference data is a statistical representation of feature data of the plurality of frames.

Abstract: A method (300) and apparatus (400) is described for segmenting an image (102). Starting with each pixel of the image (102) being a separate region, segments are formed by merging the regions. As merging proceeds, a merging cost of the regions being merged generally increases. This increase however is not purely monotonic as the overall rise in the merging cost is punctuated by departures from monotonicity. A complete pass is made through the segmentation, in which all regions are merged until only one remains. By analysing the points immediately after significant departures from monotonicity, a final segmentation stopping value (?stop) is chosen as being the last return to monotonicity from such a significant departure. Segmentation is repeated until the merging cost reaches the final segmentation stopping value (?stop).

Abstract: A method of animating a sprite in a video production comprising a plurality of sequential video frames is disclosed. The method comprises the steps of selecting (2004) a feature, with which the sprite is to be juxtaposed, in one of said video frames, applying (2006) a feature tracking process to the video production to thereby output (2008), for a series of said plurality of video frames containing the feature, a temporal-spatial record for the feature across the plurality of video frames, and compositing (2010), with the series of said plurality of video frames, a corresponding series of instantiations of the sprite dependent upon the temporal-spatial record.

Abstract: A method (500, 600) of tracking a feature (201) across a sequence of image frames is disclosed. A current feature position in a current frame is estimated from at least a previous feature position in a previous frame. Feature data is extracted from pixel data of the current frame. The feature data is then compared with reference data. If a difference between the feature data and the reference data is larger than a predetermined number, then track of the feature (201) has been lost. Alternatively, the current feature position is appended to the trajectory of the feature, and the reference data is updated periodically with feature data of a plurality of frames. Preferably the reference data is a statistical representation of feature data of the plurality of frames.