Microsoft Technology Licensing, LLC

Ji-Zheng Xu

Jian Li

Feng Wu

Ruiqin Xiong

Yu Wang

Fei Liang

Xiulian Peng

Tiantian Sun

Innovations in encoder-side decisions that use the results of hash-based block matching are presented. For example, some of the innovations relate to ways of building hash tables that include some (but not all) uniform blocks. Other innovations relate to ways of determining motion vector resolution based on results of hash-based block matching. Still other innovations relate to scene change detection, including long-term reference picture selection and picture quality determination during encoding.

Hash-based encoder decisions for video coding

Innovations in unified intra block copy ('BC') and inter prediction modes are presented. In some example implementations, bitstream syntax, semantics of syntax elements and many coding/decoding processes for inter prediction mode are reused or slightly modified to enable intra BC prediction for blocks of a frame. For example, to provide intra BC prediction for a current block of a current picture, a motion compensation process applies a motion vector that indicates a displacement within the current picture, with the current picture being used as a reference picture for the motion compensation process. With this unification of syntax, semantics and coding/decoding processes, various coding/decoding tools designed for inter prediction mode, such as advanced motion vector prediction, merge mode and skip mode, can also be applied when intra BC prediction is used, which simplifies implementation of intra BC prediction.

Unified intra block copy and inter prediction modes

Some implementations can compress a digital image to create a compressed image corresponding to the digital image. The digital image can be divided into a plurality of coding units and a determination can be made whether to divide each coding unit of the plurality of coding units into smaller coding units/prediction units based on (i) a range of pixel values in each coding unit and/or (ii) a number of bits to encode the coding unit. Rate distortion optimization can be performed based on a size of each of the prediction units. A determination can be made whether to perform a transform for each of the transform units based on a size of each prediction unit and/or a spatial frequency measure associated with each transform unit.

Video encoder for images

Video data for a high resolution image unit is coded with regard to both a low resolution reference image unit and a high resolution reference image unit. In an example encoding implementation, both low pass information and high pass information of residue data for a current image are generated. In an example decoding implementation, a current image is reconstructed by synthesizing both low pass information and high pass information for the reconstructed image.

Spatially scalable video coding

A non-dyadic spatial scalable wavelet transform may scale an original digital video frame or digital image at a non-dyadic ratio. The digital video frame or digital image is quantized to create a set of data representing the digital video frame or digital image. The set of data is then input to the non-dyadic spatial scalable wavelet transform. The non-dyadic spatial scalable wavelet transform may then transform the data associated with a first pixel to a high-pass coefficient and use the high-pass coefficient to transform the data associated with a second and third pixel to low-pass coefficients. The low-pass coefficients may then be converted to a digital image for viewing.

Non-dyadic spatial scalable wavelet transform

A 'directional lapped transform coder' ('DLT Coder') provides various techniques for designing directional lapped transforms. In general, lapped transforms are factorized into lifting steps. A &#x201c;directional operator&#x201d; is then introduced into each lifting step in order to construct the directional lapped transform by selecting data elements from a data signal along a path corresponding to the directional operator. The resulting directional lapped transform preserves the advantages of conventional lapped transforms while also providing more efficient representation directional signals. In various embodiments, the resulting directional lapped transform is used to enable an image/video coding scheme that provides significant improvement over conventional state-of-the-art image lapped transform based coding schemes for images with strong directional correlations.

Lifting-based directional lapped transforms

An exemplary method for encoding an image includes receiving image data, detecting edges in the image data, selecting at least some of the detected edges, encoding the selected edges as selected edge information, down-sampling the image data, encoding the down-sampled image as down-sampled image information and multiplexing the selected edges information and the down-sampled image information. In such a method, the selected edges information and the down-sampled image information can be stored as an encoded image file. Other exemplary methods, devices, systems are also disclosed.

Efficient image representation by edges and low-resolution signal

Systems and methods for using selective inter-layer prediction during layered video coding operations are described. In one aspect, a layered coded video structure is generated by selectively employing only objectively efficient inter-layer predictions between respective frames. Responsive to receiving a request for coded video data, one or more layers of the layered coded video structure are communicated to an application for decoding and presentation of video data to a user.

Growing community of inventors

Beijing, China

Ji-Zheng Xu