Abstract: A system and method for determining the efficient number of reference frames which are to be evaluated for an image using frame level adaptation techniques. The system and method determines and updates the number of reference frames which are to be searched based on the reference frame selection statistics of the previously coded frame. Preferably, the invention increases the number of reference frames to be referred when a first desired ratio of blocks refer to the furthest reference frame. The invention decreases the number of reference frames if a second desired ratio of blocks refer to the furthest reference frame. The motion value is considered in increasing the number of reference frames. The invention evaluates whether portions of the image frames are static, thereby terminating motion vector searches on that portion.

Abstract: A method and system for video compression prediction mode determination is disclosed. For one embodiment of the invention, the frequency domain characteristics (FDCs) of an image block are determined. The FDCs of the image block are compared to the FDCs of each of a number of prediction modes. The prediction mode having FDCs most similar to those of the image block is then determined and selected for encoding the block. For one embodiment of the invention the FDCs of the nine 4×4 prediction modes of the H.264 standard and the FDCs of a 4×4 image block are defined using a frequency domain transform. For one embodiment of the invention, one or more additional prediction modes are evaluated, and of these, the prediction mode resulting in RD optimization is selected for encoding the block.

Abstract: Systems and computer-readable medium are disclosed for automatic white balancing of an input image. Accordingly, the system includes a cast detector configured to receive and classify an input image based at least in part on a selection of constituent pixels of the input image. The cast detector is further configured to select the constituent pixels based on a computed reliability of each of the constituent pixels to convey at least one characteristic of a capture illuminant. The system further includes a cast balancer. The cast balancer is configured to perform white balancing of the input image based on the classification of the input image by the cast detector. The cast balancer is further configured to identify the capture illuminant based on at least one characteristic represented by each of the selected constituent pixels.

Abstract: Systems and computer-readable medium are disclosed for automatic white balancing of an input image. Accordingly, the system includes a cast detector configured to receive and classify an input image based at least in part on a selection of constituent pixels of the input image. The cast detector is further configured to select the constituent pixels based on a computed reliability of each of the constituent pixels to convey at least one characteristic of a capture illuminant. The system further includes a cast balancer. The cast balancer is configured to perform white balancing of the input image based on the classification of the input image by the cast detector. The cast balancer is further configured to identify the capture illuminant based on at least one characteristic represented by each of the selected constituent pixels.

Abstract: In one example embodiment, a method enables, computing, for a first pixel in the digital mosaiced image, the first pixel being characterized by a first color component and a first set of gradient values in a plurality of orientations, gradient values in the plurality of orientations for a second pixel in the neighborhood of the first pixel. Color values in the plurality of orientations corresponding to a second color component associated with the first pixel based on the set of first gradient values are estimated. The first set of gradient values based at least in part on the computed gradient values is updated. One of the plurality of orientations of the estimated color value based on the updated set of first gradient values is selected and one of the estimated color values corresponding to the selected orientation is determined.

Abstract: A method and system for video compression prediction mode determination is disclosed. For one embodiment of the invention, the frequency domain characteristics (FDCs) of an image block are determined. The FDCs of the image block are compared to the FDCs of each of a number of prediction modes. The prediction mode having FDCs most similar to those of the image block is then determined and selected for encoding the block. For one embodiment of the invention the FDCs of the nine 4×4 prediction modes of the H.264 standard and the FDCs of a 4×4 image block are defined using a frequency domain transform. For one embodiment of the invention, one or more additional prediction modes are evaluated, and of these, the prediction mode resulting in RD optimization is selected for encoding the block.

Abstract: The fact that fractional-pixel points that are searched are close to each other, and hence the half-pixel and quarter-pixel interpolation of each point involves computing interpolations that have already been computed, may be exploited. This is accomplished by storing the interpolated values of the first point and reusing them when computing the interpolations for the other points. This may be performed for both horizontal and vertical interpolations. For interpolation in both the horizontal and vertical directions, the horizontal interpolated values may be reused, and vertical interpolation carried out over them, or vice-versa. For computing the quarter-pixel interpolated values, the half-pixel interpolated values computed and stored previously may be used.