Canon Kabushiki Kaisha

Ka-Ming Leung

Axel Lakus-Becker

Zhonghua Ma

Timothy Merrick Long

Simon John Liddington

A system () for encoding an input video frame (), for transmitting or storing the encoded video and for decoding the video is disclosed. The system () includes an encoder () and a decoder () interconnected through a storage or transmission medium (). The encoder () includes a turbo encoder () for forming parity bit data from the input frame () into a first data source (), and a sampler () for down-sampling the input frame () followed by intraframe compression () to form a second data source (). The decoder () receives data from the second data source () to form an estimate for the frame (). The decoder () also receivers the parity bit data from the first data source (), and corrects errors in the estimate by applying the parity bit data to the estimate. Each bit plane is corrected in turn by a turbo decoder (). The decoder determines how reliably a pixel value was decoded, too. Frame reconstruction module () takes advantage of this and discards unreliably decoded pixels and replaces them with predicted pixel values.

Wyner ziv coding

A method of selecting a first image from a plurality of images for constructing a coordinate system of an augmented reality system. A first image feature in the first image corresponding to the feature of the marker is determined. A second image feature in a second image is determined based on a second pose of a camera, said second image feature having a visual match to the first image feature. A reconstructed position of the feature of the marker in a three-dimensional (3D) space is determined based on positions of the first and second image features, the first and the second camera pose. A reconstruction error is determined based on the reconstructed position of the feature of the marker and a pre-determined position of the marker.

Key-frame selection for parallel tracking and mapping

A method () of performing distributed video encoding on an input video frame (), is disclosed. The method () forms a bit-stream from original pixel values of the input video frame (), such that groups of bits in the bit-stream are associated with clusters of spatial pixel positions in the input video frame (). The bit-stream is interleaved to reduce the clustering. The interleaved bit-stream is encoded to generate parity bits from the bit-stream according to a bitwise error correction method.

For error correction in distributed video coding

A method of determining bit rates for use in encoding video data for joint decoding, is disclosed. An approximation of the video data is generated for later use as side information during a process of joint decoding. Bit error probabilities are determined for each bit plane and for each coefficient band of the approximation. The bit rates are determined for encoding the bit planes depending on the bit error probabilities, bit planes, and coefficient bands.

Rate-distortion control in DVC with no feedback channel

Disclosed is a method of image coding for joint decoding of images from different viewpoints using distributed coding techniques. The method receives a first set of features () and error correction bits () corresponding to a first image () obtained at a first viewpoint () and a second set of features () from a second image () corresponding to a second viewpoint (). An approximation () of said first image () at said first viewpoint () is determined () an based on the first and second sets of features () and the second image at the second viewpoint. A reliability measure () of the approximation of the first image is then determined () by joint decoding () the approximation () using the error correction bits (). The approximation of the first image is then refined iteratively () based on the reliability measure () and image information () derived from the joint decoding.

Robust image alignment for distributed multi-view imaging systems

A method of encoding video data generates a first source of video data from a first set of video frames by approximating the first set of video frames. A second source of video data is generated from a second set of video frames by transforming first respective binary representations of pixel values of said second set of video frames into second respective binary representation of the pixel values of the second set of video frames. The video data sources are encoded independently according to a mapping wherein Hamming distance between each of successive pixel values in a predetermined range of values in the second binary representation is greater than Hamming distance between each of successive pixel values in a predetermined range of values in the first binary representation.

Spatial Wyner Ziv coding

Methods, apparatuses (), and computer program products for generating an enhanced digital image () comprising a plurality of pixels are disclosed. Using a first digital image () captured from a first camera () and parity bits () generated from a second digital image captured by a second camera (), a third digital image () is constructed. The second camera () captures the second image at a resolution different to the resolution of the first camera () capturing the first image (). A disparity map () between the first image () and the third image () is determined (). One of the first image () and the third image () is enhanced () dependent upon the determined disparity map () to generate the enhanced digital image ().

Growing community of inventors

Marsfield, Australia

Ka-Ming Leung