A method for reducing color leaking artefacts in an image formed by projection processing from a 3D point cloud comprises: receiving an input image comprising the 3D point cloud; classifying the cloud into a plurality of surface patches; projecting the patches onto a plane to form a first 2D image; processing the first 2D image, by coding, transmitting and decoding, to form a final 2D image; and providing the final 2D image as an output. Processing includes independent patch processing to reduce inter-patch color leakage in the final 2D image, the independent patch processing including chroma sub-sampling pixels within each of the projected patches in the first 2D image separately; recombining the chroma sub-sampled patches to form a second 2D image; and compressing the second 2D image.

A method of encoding video data, the method comprising: generating prediction data of the video data; generating residual data based on the prediction data and digital sample values of the video data; generating coefficients based on the residual data; performing an interlacing process to generate an interlaced amplitude value, wherein the interlacing process interlaces bits of two or more of the coefficients to generate the interlaced amplitude value; modulating an analog signal based on the interlaced amplitude value; generating digital values based on the prediction data; and outputting the analog signal and digital values based on the prediction block.

Systems and methods are disclosed for image signal processing. For example, methods may include receiving a current image of a sequence of images from an image sensor; combining the current image with a recirculated image to obtain a noise reduced image, where the recirculated image is based on one or more previous images of the sequence of images from the image sensor; determining a noise map for the noise reduced image, where the noise map is determined based on estimates of noise levels for pixels in the current image, a noise map for the recirculated image, and a set of mixing weights; recirculating the noise map with the noise reduced image to combine the noise reduced image with a next image of the sequence of images from the image sensor; and storing, displaying, or transmitting an output image that is based on the noise reduced image.

Systems and methods are disclosed for image signal processing. For example, methods may include receiving a current image of a sequence of images from an image sensor; combining the current image with a recirculated image to obtain a noise reduced image, where the recirculated image is based on one or more previous images of the sequence of images from the image sensor; determining a noise map for the noise reduced image, where the noise map is determined based on estimates of noise levels for pixels in the current image, a noise map for the recirculated image, and a set of mixing weights; recirculating the noise map with the noise reduced image to combine the noise reduced image with a next image of the sequence of images from the image sensor; and storing, displaying, or transmitting an output image that is based on the noise reduced image.

An encoding method, a decoding method, and an apparatus are provided. The encoding method includes: determining X first patches from M first patches in a current frame, where there is a matching relationship between the X first patches and X second patches, the X second patches are included in a previous frame of the current frame, X is less than or equal to M, and both X and M are positive integers; obtaining auxiliary information of the X first patches and auxiliary information of X second patches; obtaining X groups of auxiliary information differences based on the auxiliary information of the X first patches and the auxiliary information of the X second patches; and encoding the X groups of auxiliary information differences. The encoding method improves coding performance by using a correlation between point cloud data of two adjacent frame.

An encoding method, a decoding method, and an apparatus are provided. The encoding method includes: determining X first patches from M first patches in a current frame, where there is a matching relationship between the X first patches and X second patches, the X second patches are included in a previous frame of the current frame, X is less than or equal to M, and both X and M are positive integers; obtaining auxiliary information of the X first patches and auxiliary information of X second patches; obtaining X groups of auxiliary information differences based on the auxiliary information of the X first patches and the auxiliary information of the X second patches; and encoding the X groups of auxiliary information differences. The encoding method improves coding performance by using a correlation between point cloud data of two adjacent frame.

Certain aspects of the present disclosure are directed to methods and apparatus for compressing video content using deep generative models. One example method generally includes receiving video content for compression. The received video content is generally encoded into a latent code space through an auto-encoder, which may be implemented by a first artificial neural network. A compressed version of the encoded video content is generally generated through a trained probabilistic model, which may be implemented by a second artificial neural network, and output for transmission.

Aspects of the disclosure provide methods and apparatuses for point cloud compression. In some examples, an apparatus for point cloud compression includes processing circuitry. In some embodiments, the processing circuitry determines one or more original points in a point cloud that are associated with a reconstructed position. Positions of the one or more original points can be reconstructed, according to a geometry quantization, to the reconstructed position. The processing circuitry then determines an attribute value for the reconstructed position based on attribute information of the one or more original points, and encodes texture of the point cloud with the reconstructed position having the determined attribute value.

A system and method for lossy image and video compression that utilizes a metanetwork to generate a set of hyperparameters necessary for an image encoding network to reconstruct the desired image from a given noise image, and for lossy image and video compression and transmission that utilizes a neural network as a function to map a known noise image to a desired or target image, allowing the transfer only of hyperparameters of the function instead of a compressed version of the image itself. This allows the recreation of a high-quality approximation of the desired image by any system receiving the hyperparameters, provided that the receiving system possesses the same noise image and a similar neural network. The amount of data required to transfer an image of a given quality is dramatically reduced versus existing image compression technology.

A compression system trains a machine-learned compression model that includes components for an encoder and decoder. In one embodiment, the compression model is trained to receive parameter information on how a target frame should be encoded with respect to one or more encoding parameters, and encodes the target frame according to the respective values of the encoding parameters for the target frame. In particular, the encoder of the compression model includes at least an encoding system configured to encode a target frame and generate compressed code that can be transmitted by, for example, a sender system to a receiver system. The decoder of the compression model includes a decoding system trained in conjunction with the encoding system. The decoding system is configured to receive the compressed code for the target frame and reconstruct the target frame for the receiver system.