Telefonaktiebolaget Lm Ericsson (publ)

Julien Michot

Kenneth Andersson

Thomas Rusert

Jack Enhorn

It is presented a method for encoding a video frame of a piece of video content. The method is performed in an encoder device and comprises the steps of: dividing the video frame into a set of non-overlapping regions, wherein the set comprises at least one region; assigning each region, in the set of regions, to a separate processing unit; encoding, in each processing unit, picture data of the respective region; collecting progress data from each processing unit, the progress data indicating progress of the encoding of the respective region; and dividing a particular region into a plurality of smaller regions and assigning each smaller region to a separate processing unit, when the progress data indicates that progress of encoding the particular region is insufficient.

Encoding and decoding a video frame in separate processing units

A method of generating a depth cue for three dimensional video content is disclosed. The method comprises the steps of (a) detecting three dimensional video content that will appear in observer space when displayed () (b) identifying a reference projection parameter () (c) estimating a location of a shadow that would be generated by the detected content as a consequence of a light source emitting light according to the reference projection parameter () and (d) projecting light content imitating a shadow to the estimated location to coincide with display of the three dimensional video content (). Also disclosed are a computer program product for carrying out a method of generating a depth cue for three dimensional video content and an apparatus () for generating a depth cue for three dimensional video content.

Method and apparatus for generating a depth cue

According to embodiments of the present invention, sound localization is used to determine the active speaker in a video conference. A network element uses the localization to determine which regions of the image that should be preserved and retargets the video accordingly. By providing a retargeted video where the speaker is more visible, a better user experience is achieved.

Arrangements and method thereof for video retargeting for video conferencing

A method for determining a depth of a target object in an environment is disclosed. The method comprises sensing a depth of objects within the environment with a depth sensing apparatus (), identifying a disocclusion gap caused by the target object (), detecting a position of edge contours of the disocclusion gap (), and calculating a depth of the target object from the position of the disocclusion gap edge contours and a position of the depth sensing apparatus (). Also disclosed are a computer program product for carrying out a method for determining a depth of a target object in an environment and an apparatus () for determining a depth of a target object in an environment.

Growing community of inventors

Sundbyberg, Sweden

Julien Michot