Abstract: A vertical scrolling region detector may include a motion estimator that can generate motion vectors between blocks of two or more primary frames of a mixed-mode video sequence. The detector may also include a primary frame motion analyzer that can analyze the motion vectors to detect substantially constant vertical motion of at least some of the blocks between the two or more primary frames. The presence of substantially constant vertical motion may reflect the presence of a vertical scrolling region in the mixed-mode video sequence. Moreover, the detector may also include a consecutive frame motion analyzer that can calculate differences in pixel values between lines of two or more consecutive frames in the mixed-mode video sequence. The differences in pixel values may further reflect the presence of the vertical scrolling region in the mixed-mode video sequence.

Abstract: Embodiments include systems and methods of determining block edge location in video. In one embodiment, a system includes a block edge map generator configured to generate a binary block map, the block map having indicators representing the locations of block edges in a first direction, a histogram generator configured to accumulate the indicators to form a block edge histogram, the peaks of the histogram corresponding to block edge locations in the image, a low-pass filter to low-pass filter the block edge histogram and form a filtered histogram, a peak locator configured to identify the peaks of the filtered histogram, and a peak refiner configured to refine locations of peaks located in the filtered block edge histogram based on the identified peaks and peaks in the block edge histogram, the refined peak locations indicative of block edge locations, and further configured to provide block locations.

Abstract: A system and method for adaptively defining a region of interest for motion analysis in digital video is disclosed. In one embodiment, a method of detecting a gesture is disclosed which comprises receiving a video sequence comprising a plurality of frames, determining a region of interest which excludes a portion of the frame, and detecting the gesture within the region of interest.

Abstract: A general shoot suppression method and system for image detail enhancement that provides good suppression in shoot areas while preserving the enhancement (with less or no suppression) in non-shoot areas. Asymmetry checker checks the luminance variation of pixels around the current pixel within a filtering range. Based on the checking result, different patterns of luminance variation in the neighborhood of the current pixel are categorized. Higher suppression is applied to those patterns that are more likely to cause overshoot/undershoot. Optionally, suppression is further associated with the intensity of detail signal at each pixel position. An intensity checker checks the magnitude of the detail signal. If the detail signal is weak at a current pixel position, it is unlikely that obvious shoot artifacts appear at that position, and little or no suppression is applied. Such a method and system provide shoot suppression only in shoot areas while maintaining good enhancement in non-shoot areas.

Abstract: A method for reliability estimation of temporal noise estimation in a sequence of video frames. The temporal local differences from a difference between a previous frame and a next frame in the sequence of frames is determined. A distribution of the temporal local difference is determined. Characteristics values of the distribution are determined. The characteristics values are compared to the thresholds to obtain an indication of the reliability of the temporal noise estimation. If the estimated noise variance is determined as not reliable, it will be discarded and the previous estimated reliable noise variance will be used instead indicating the noise level of the current frame.

Abstract: A noise estimation system that estimates noises in a sequence of video frames, uses a an image structure remover and a noise variance calculator. The image structure remover computes the difference over a local window in two consecutive frames. The noise variance calculator estimates the noise variance (standard deviation) from the distribution of the local difference. If there is no motion or small motion between two consecutive frames, the image structure can be removed by calculating the local difference, whereby very robust estimation is obtained.

Abstract: An estimation system provides quality measure of a given image for image/video applications to determine if the image needs to be enhanced or not. The system systemically measures the image quality by estimating quantization parameter of the compressed digital image. This allows estimating the image quantization parameter effectively in the pixel domain. Because block artifact are caused by image compression transforms, especially in flat image region, the estimation system first detects the suspicious regions of the image for the block artifacts by calculating a variance for each location in the image. Next, noise power is computed by using the detected regions. Then, the estimated noise power is compared with the statistical model to calculate the quantization parameter that is used in the encoder side when the input image is compressed. Finally, a quantization parameter is used as a quality measure of the input image.

Abstract: A method of image restoration in an image processing system is provided. An observed image of an original image, wherein the observed image comprises multiple samples, is received. For a selected sample from the multiple samples, a subset of the multiple samples is selected, wherein said subset includes the selected sample. Image restoration to restore an original image sample from the selected sample is performed as a function of said subset of multiple samples, and not the entire observed image. A novel image restoration method which requires shorter length FFT calculation in the image restoration process is utilized, wherein the resulting restored image maintains the same quality as the result from conventional approaches.

Abstract: Embodiments include systems and methods for interpolating frames. In one embodiment, a system and method provides reduced halo and consequently a higher degree of accuracy/fidelity. In one embodiment, a method of video frame interpolation includes receiving forward and backward motion estimation data based on a first video frame and a second, subsequent, video frame. Each block in a plurality of blocks of pixels in an intermediate frame between the first and second frames is classified according to measures of covering and uncovering associated with the forward and backward motion estimation data. A block to be constructed is estimated based on the classification results of proximal ones of the plurality of blocks to the block to be constructed.

Abstract: An image signal processing unit and an image signal processing method thereof An image signal processing unit estimates motion characteristics of an input image, and adjusts a phase of an interpolation frame according to the motion characteristics when converting a frame rate. Accordingly, linearity can be maintained according to the motion characteristics of an input image, and the frame rate can be converted without generating noise such as halo artifacts.

Abstract: A system and method for controlling a device via gesture recognition is disclosed. In one embodiment, the system comprises a video capture device configured to capture video of an object, a tracking module configured to track the position of the object, thereby defining a trajectory, a trajectory analysis module configured to determine whether or not a portion of the trajectory defines a recognized gesture, and a control module configured to change a parameter of the device when it is determined that the trajectory of the object defines a recognized gesture.

Abstract: A ringing area detector classifies the input image into two regions: a mosquito noise region (i.e. filtering region) and a non-mosquito noise region (i.e. non-filtering region), and uses this classification information to adaptively remove the mosquito noise in a mosquito noise reduction system. The mosquito noise reduction system includes a ringing area detector, a local noise power estimator, a smoothing filter, and a mixer. The ringing area detector includes an edge detector, a near edge detector, a texture detector, and a filtering region decision block. The ringing detection block detects the ringing area where the smoothing filter is to be applied. The local noise power estimator controls the filter strength of the smoothing filter. The smoothing filter smoothes the input image. The mixer mixes the smoothed image and the original image properly based on the region information from the ringing area detection block.

Abstract: A method of computing a motion decision signal for a video processing system, by: inputting a video signal including a video sequence of frames; filtering a frame in the video signal; computing a point-wise temporal local difference signal between the filtered frame and a next frame in the video sequence; computing a point-wise motion decision signal based on the temporal local difference signal.

Abstract: A method and system for generating intermediate fields of a video signal are provided. One implementation involves generating intermediate fields of a video signal, by performing interpolation for generating said intermediate (missing) fields of said video signal; and applying fallback interpolation when quality of the interpolated fields is below a threshold or when valid motion information cannot be determined. Fallback interpolation involves estimating a global motion of a scene represented by the video signal, determining a periodic motion count based on the estimated global motion, and controlling application of fallback interpolation based on said periodic motion count.

Abstract: A method and system for detecting motion at a temporal intermediate position between image fields is provided. One implementation involves detecting an uncovering area in the temporal intermediate position in an image field; determining a motion vector candidate, in place of a current original motion vector, for the temporal intermediate position with a detected uncovering area; and determining a motion vector representing motion at the temporal intermediate position by combining the candidate motion vector with a current original motion vector for the temporal intermediate position. Erroneous motion vectors in uncovering areas are eliminated.

Abstract: A system and method of providing motion estimation is disclosed. A recursive motion estimation module applies multiple recursive motion estimation rules which have complimentary performance advantages. A selection module provides a switching scheme which determines which rule to apply to a specific block within a video frame.

Abstract: A method and apparatus for detecting and locating objects of interest in video sequences is provided. A frame is defined as an image belonging to video sequences. Each frame with the same or different size of original input sequences is searched by the same or different size window efficiently for detecting objects. The characteristics of temporal redundancies in video sequences are used in detecting objects in video sequences.

Abstract: Systems and methods for estimating the centers of moving objects in a video sequence are disclose. One embodiment is a method of defining one or more motion centers in a video sequence, the method comprising receiving a video sequence comprising a plurality of frames, receiving a motion history image for each of a subset of the plurality of frames based on the video sequence, identifying, through use of the motion history image, one or more data segments having a first orientation, wherein each data segment having the first orientation has a start location and a length, identifying, one or more data segments having a second orientation, wherein each element of a data segment having the second orientation is associated with a data segment having the first orientation, and defining a corresponding motion center for one or more of the indentified data segments having the second orientation.

Abstract: An image processing system provides image enhancement and anti-clipping units. The anti-clipping unit for image sharpness enhancement, operates such that any shoot artifacts in the enhanced image that go beyond pixel value lower/upper bounds are properly adjusted back within the lower and upper bounds, without causing prominent edge jaggedness artifacts in the final resulting output image.

Abstract: A method for adaptive reduction of ringing artifacts in an input image including pixels of image information. The method includes selecting a pixel window including a set of pixels from the input image pixels, detecting areas of ringing artifacts in the pixel window based on the pixel information, processing the pixels in the detected areas to reduce the detected ringing artifacts in those areas, and generating an enhanced output image including the processed pixels with reduced ringing artifacts.