Abstract: One embodiment provides a method comprising receiving content for presentation on a display, and obtaining a plurality of image patches from the content. Each image patch is segmented from a bottom and horizontal-center portion of a sub-sampled grayscale image of a frame of the content. The method further comprises applying a searching algorithm to the plurality of image patches to detect a region of interest of moving text in the content. The searching algorithm utilizes a first set of features and a second set of features to detect the region of interest with respect to a vertical axis and a horizontal axis of the display, respectively. The method further comprises correcting one or more visual artifacts in the region of interest, where the one or more visual artifacts include broken text.

Abstract: One embodiment provides a method comprising receiving content for presentation on a display, and obtaining one or more sample frames of the content. The method further comprises generating a set of features based on one or more horizontal edge signals and one or more vertical edge signals of the one or more sample frames. The method further comprises utilizing a classification model to detect, based on the set of features, a region of interest of moving text in the one or more sample frames.

Abstract: One embodiment provides a computer-implemented method that includes comparing, by a computing device, an input character signal with one or more prestored character templates for determining an estimated difference measure. The computing device, based on the estimated difference measure, determines one or more mixing weights between a stored character patch buffer and a current input character patch for determining an output mixing patch. The computing device further updates the character patch buffer based on the output mixing patch. The computing device additionally substitutes a designated area using the output mixing patch to produce a final output.

Abstract: One embodiment provides a computer-implemented method that includes receiving content having a first resolution. The processor splits the content into most significant bits and least significant bits. The processor further performs processing of the most significant bits of the content while preserving the least significant bits. The processor additionally performs a linear combination of outputs of the processing of the most significant bits with the preserved least significant bits for improving quality of the content.

Abstract: One embodiment provides a method comprising segmenting an input video into a first video chunk and one or more subsequent video chunks. The method further comprises, for each subsequent video chunk, generating a corresponding window of overlapped frames by selecting a subsequence of frames from a different video chunk immediately preceding the subsequent video chunk. The method further comprises generating metadata corresponding to each video chunk by processing each video chunk in parallel. Each subsequent video chunk is processed based in part on a corresponding window of overlapped frames. The method further comprises, for each subsequent video chunk, discarding a portion of metadata corresponding to the subsequent video chunk, where the portion discarded is specific to a corresponding window of overlapped frames. The method further comprises merging each video chunk into a single output video. Each video chunk merged is associated with any remaining corresponding metadata.

Abstract: One embodiment provides a method comprising acquiring 3D content comprising a 3D object in 3D space. The 3D object has object information indicative of a location of the 3D object in the 3D space. The method further comprises projecting the 3D object to a 2D object in 2D space based on the object information. The 2D object has one or more 2D vertices indicative of a location of the 2D object in the 2D space. The method further comprises determining one or more latent variables in the 2D space based on the object information and the one or more 2D vertices, and generating a 2D feature database including the one or more latent variables.

Abstract: One embodiment provides a method comprising segmenting an input video into a first video chunk and one or more subsequent video chunks. The method further comprises, for each subsequent video chunk, generating a corresponding window of overlapped frames by selecting a subsequence of frames from a different video chunk immediately preceding the subsequent video chunk. The method further comprises generating metadata corresponding to each video chunk by processing each video chunk in parallel. Each subsequent video chunk is processed based in part on a corresponding window of overlapped frames. The method further comprises, for each subsequent video chunk, discarding a portion of metadata corresponding to the subsequent video chunk, where the portion discarded is specific to a corresponding window of overlapped frames. The method further comprises merging each video chunk into a single output video. Each video chunk merged is associated with any remaining corresponding metadata.

Abstract: One embodiment provides a method comprising generating, via an edge detection algorithm, a first edge map based on a first frame of an input content comprising a sequence of frames. The method further comprises generating, via the edge detection algorithm, a second edge map based on a second frame of the input content. The first frame precedes the second frame in the sequence of frames. The method further comprises determining a difference between the first edge map and the second edge map, determining a metric indicative of an estimated amount of judder present in the input content based on the difference, and dynamically adjusting a frame rate of the input content based on the metric. The input content is displayed on a display device at the adjusted frame rate.

Abstract: A method, apparatus, and computer readable medium for frame rate conversion based on object tracking. The method includes determining whether a block of a first frame of an input video stream corresponds to a detected object region or a non-object region. The method further includes selecting a motion estimation operation from a group of operations comprising an object tracking operation based at least on the determining of whether the block of the first frame corresponds to the detected object region or the non-object region. The method further includes determining a motion vector of the block by performing the selected motion estimation operation and determining an interpolated frame based at least on the determined motion vector and the first frame and a second frame of the input video stream.

Abstract: A method of operation of a depth estimation system includes: calculating focus measures for positions on a two-dimensional image; generating a depth map for the two-dimensional image based on fitting the focus measure through a Gaussian function; generating a three-dimensional image from the depth map and the two-dimensional image; and processing the three-dimensional image on a storage unit for displaying on a device.

Abstract: A method of operation of a display system includes: calculating a focus measure for an original image; calculating a segment mean based on the focus measure for a segment; generating an ordered segment based on the segment mean; generating a segment depth based on the ordered segment; and generating a three-dimensional image with the segment depth for displaying on a device.

Abstract: A method of operation of a display system includes: calculating a focus measure for an original image; calculating a segment mean based on the focus measure for a segment; generating an ordered segment based on the segment mean; generating a segment depth based on the ordered segment; and generating a three-dimensional image with the segment depth for displaying on a device.

Abstract: A method of operation of a depth estimation system includes: calculating focus measures for positions on a two-dimensional image; generating a depth map for the two-dimensional image based on fitting the focus measure through a Gaussian function; generating a three-dimensional image from the depth map and the two-dimensional image; and processing the three-dimensional image on a storage unit for displaying on a device.