4604 results about "Video sequence" patented technology

Disclosed is a method and system for efficiently and accurately tracking three-dimensional (3D) human motion from a two-dimensional (2D) video sequence, even when self-occlusion, motion blur and large limb movements occur. In an offline learning stage, 3D motion capture data is acquired and a prediction model is generated based on the learned motions. A mixture of factor analyzers acts as local dimensionality reducers. Clusters of factor analyzers formed within a globally coordinated low-dimensional space makes it possible to perform multiple hypothesis tracking based on the distribution modes. In the online tracking stage, 3D tracking is performed without requiring any special equipment, clothing, or markers. Instead, motion is tracked in the dimensionality reduced state based on a monocular video sequence.

A machine includes a processor and a memory connected to the processor. The memory stores instructions executed by the processor to receive a message with audio content and video content. Audio fingerprints within the audio content are evaluated. The audio fingerprints are matched to known audio fingerprints to establish matched audio fingerprints. A determination is made whether the matched audio fingerprints correspond to a designated gallery constructed to receive a sequence of videos set to an audio time line. The matched audio fingerprints and corresponding video content are added to the audio time line. The operations are repeated until the audio time line is populated with corresponding video content to form a completed gallery with video segments set to audio segments that constitute a complete audio time line. The completed gallery is supplied in response to a request.

The invention relates to a method and a graphical user interface for modifying a depth map for a digital monoscopic color image. The method includes interactively selecting a region of the depth map based on color of a target region in the color image, and modifying depth values in the thereby selected region of the depth map using a depth modification rule. The color-based pixel selection rules for the depth map and the depth modification rule selected based on one color image from a video sequence may be saved and applied to automatically modify depths maps of other color images from the same sequence.

There is disclosed a method and device of statistical interleaving of at least two digital video sequences each comprising a plurality of coded pictures to be reproduced by a decoder, wherein the two digital video sequences form a single video stream and are destined for being successively reproduced in time, comprising statistical multiplexing a first digital video sequence with a second digital video sequence, wherein the pictures of the second digital video sequence are associated with timing information, which prevents a decoder from reproducing the pictures; transmitting the digital video sequences in an interleaved manner in accordance with a result of the statistical multiplexing to the decoder; and once the second digital video sequence is to be reproduced by decoder receiving the stream: transmitting dummy pictures, which are coded to refer to one or more pictures of the second video sequence transmitted in advance to the decoder.

The image generation apparatus includes: an image receiving unit which receives a first video sequence including frames having a first resolution and a second video sequence including frames having a second resolution which is higher than the first resolution, each frame of the first video sequence being obtained with a first exposure time, and each frame of the second video sequence being obtained with a second exposure time which is longer than the first exposure time; and an image integration unit which generates, from the first video sequence and the second video sequence, a new video sequence including frames having a resolution which is equal to or higher than the second resolution, at a frame rate which is equal to or higher than a frame rate of the first video sequence, by reducing a difference between a value of each frame of the second video sequence and a sum of values of frames of the new video sequence which are included within an exposure period of the frame of the second video sequence.

A more efficient way of addressing different bit-depths, or different bit-depths and chroma sampling format requirements is achieved by using a low bit-depth and / or low-chroma resolution representation for providing a respective base layer data stream representing this low bit-depth and / or low-chroma resolution representation as well as for providing a higher bit-depth and / or higher chroma resolution representation so that a respective prediction residual may be encoded in order to obtain a higher bit-depth and / or higher chroma resolution representation. By this measure, an encoder is enabled to store a base-quality representation of a picture or a video sequence, which can be decoded by any legacy decoder or video decoder, together with an enhancement signal for higher bit-depth and / or reduced chroma sub-sampling, which may be ignored by legacy decoders or video decoders.

A system and method for generating CT-type three dimensional imaging information from a conventional C-arm fluoroscope system. This enables the adaptation of widely used C-arm installations to provide CT-type information. The system uses a three dimensional target disposing in a fixed position relative to the subject, and obtains a sequence of video images of a region of interest of a subject while the C-arm is moved manually or by a scanning motor. Images from the video sequence are analyzed to determine the pose of the C-arm relative to the subject by analysis of the image patterns of the target. Images are selected from the video sequence according to predetermined criteria. A set of two-dimensional image data with associated positional data is obtained, which is used to reconstruct a three dimensional volumetric set of imaging data of the region of interest of the subject.

An object in a video sequence is tracked by object masks generated for frames in the sequence. Macroblocks are motion compensated to predict the new object mask. Large differences between the next frame and the current frame detect suspect regions that may be obscured in the next frame. The motion vectors in the object are clustered using a K-means algorithm. The cluster centroid motion vectors are compared to an average motion vector of each suspect region. When the motion differences are small, the suspect region is considered part of the object and removed from the object mask as an occlusion. Large differences between the prior frame and the current frame detect suspected newly-uncovered regions. The average motion vector of each suspect region is compared to cluster centroid motion vectors. When the motion differences are small, the suspect region is added to the object mask as a disocclusion.

Motion estimation is the science of predicting the current frame in a video sequence from the past frame (or frames), by slicing it into rectangular blocks of pixels, and matching these to past such blocks. The displacement in the spatial position of the block in the current frame with respect to the past frame is called the motion vector. This method of temporally decorrelating the video sequence by finding the best matching blocks from past reference frames—motion estimation—makes up about 80% or more of the computation in a video encoder. That is, it is enormously expensive, and methods do so that are efficient are in high demand. Thus the field of motion estimation within video coding is rich in the breadth and diversity of approaches that have been put forward. Yet it is often the simplest methods that are the most effective. So it is in this case. While it is well-known that a full search over all possible positions within a fixed window is an optimal method in terms of performance, it is generally prohibitive in computation. In this patent disclosure, we define an efficient, new method of searching only a very sparse subset of possible displacement positions (or motion vectors) among all possible ones, to see if we can get a good enough match, and terminate early. This set of sparse subset of motion vectors is preselected, using a priori knowledge and extensive testing on video sequences, so that these “predictors” for the motion vector are essentially magic. The art of this method is the preselection of excellent sparse subsets of vectors, the smart thresholds for acceptance or rejection, and even in the order of the testing prior to decision.