Visual attention model
AssigneeTektronix, Inc.
InventorWilfried M. Osberger
Life of this patent
Estimating the term…
Abstract
An improved visual attention model uses a robust adaptive segmentation algorithm to divide a current frame of a video sequence into a plurality of regions based upon both color and luminance, with each region being processed in parallel by a plurality of spatial feature algorithms including color and skin to produce respective spatial importance maps. The current frame and a previous frame are also processed to produce motion vectors for each block of the current frame, the motion vectors being compensated for camera motion, and the compensated motion vectors being converted to produce a temporal importance map. The spatial and temporal importance maps are combined using weighting based upon eye movement studies.
Claims20
An improved visual attention model of the type that segments a frame of a video sequence into regions for processing by a plurality of spatial features to produce a corresponding plurality of spatial importance maps, that compares the frame with a previous frame for processing to produce a temporal importance map, and that combines the spatial and temporal importance maps to produce a total importance map for the frame, wherein the improvement comprises the steps of:
- adaptively segmenting the frame into the regions using color along with luminance;
- processing the regions with a plurality of spatial features to produce the plurality of spatial importance maps;
- processing the frame with the previous frame to produce the temporal importance map that is compensated for camera motion;
- and
- combining the spatial and temporal importance maps based upon a weighting function derived from eye movement studies to produce the total importance map for the frame.
Dependencies are inferred from the claim text.