Abstract: A projection system and calibration method therefore relate to a light source configured to emit a light in response to an image data, an illumination optical system configured to steer the light, the illumination optical system including a fold mirror and an integrating rod, a digital micromirror device (DMD) including a plurality of micromirrors respectively configured to reflect the steered light to a predetermined location as on-state light or to reflect the steered light as off-state light to a light dump; determining a deviation between an actual angle of orientation and an expected angle of orientation of a respective micromirror of the plurality of micromirrors; calculating a first amount of rotational adjustment corresponding to the fold mirror and a second amount of lateral adjustment corresponding to the integrating rod, and actuating the fold minor and integrating rod according to the corresponding first and second amount.

Abstract: A neural network system for predicting frequency coefficients of a media signal, the neural network system comprising a time predicting portion including at least one neural network trained to predict a first set of output variables representing a specific frequency band of a current time frame given coefficients of one or several previous time frames, and a frequency predicting portion including a at least one neural network trained to predict a second set of output variables representing a specific frequency band given coefficients of one or several frequency bands adjacent to the specific frequency band in said current time frame. Such a neural network system forms a predictor capable of capturing both temporal and frequency dependencies occurring in time-frequency tiles of a media signal.

Abstract: A first predictor is applied to an input image to generate first-stage predicted codewords approximating prediction target codewords of a prediction target image. Second-stage prediction target values are created by performing an inverse cascade operation on the prediction target codewords and the first-stage predicted codewords. A second predictor is applied to the input image to generate second-stage predicted values approximating the second-stage prediction target values. Multiple sets of cascade prediction coefficients are generated to comprise first and second sets of cascade prediction coefficients specifying the first and second predictors. The multiple sets of cascade prediction coefficients are encoded, in a video signal, as image metadata. The video signal is further encoded with the input image.

Abstract: In some embodiments, an echo cancellation method which includes adaptation of at least one prediction filter, with adaptation step size controlled using gradient descent on a set of filter coefficients of the filter, where control of the adaptation step size is based at least in part on a direction of adaptation and a predictability of a gradient of adaptation (e.g., a gradient vector). Other aspects of embodiments of the invention include systems, methods, and computer program products for controlling adaptation step size of adaptive (e.g., low-complexity adaptive) echo cancellation. In some embodiments, adaptation step size control is based on a normalized, scaled gradient of adaptation, or includes smoothing of a normalized gradient of adaptation.

Abstract: Methods and systems for chromatic ambient-light support are provided. Given an input surround correlated color temperature (CCT), its normalized value is mapped to a preferred gray CCT value using a sigmoid-like function model generated based on viewer experimental data. The preferred gray CCT value is then mapped to an adjusted CCT value so that viewing conditions on a display with the input surround CCT value match a D65 surround CCT value.

Abstract: Dual and multi-modulator projector display systems and techniques are disclosed. In one embodiment, a projector display system comprises a light source; a controller, a first modulator, receiving light from the light source and rendering a halftone image of said the input image; a blurring optical system that blurs said halftone image with a Point Spread Function (PSF); and a second modulator receiving the blurred halftone image and rendering a pulse width modulated image which may be projected to form the desired screen image. Systems and techniques for forming a binary halftone image from input image, correcting for misalignment between the first and second modulators and calibrating the projector system—e.g. over time—for continuous image improvement are also disclosed.

Abstract: A method for performing DRC on a HOA signal comprises transforming the HOA signal to the spatial domain, analyzing the transformed HOA signal, and obtaining, from results of said analyzing, gain factors that are usable for dynamic compression. The gain factors can be transmitted together with the HOA signal. When applying the DRC, the HOA signal is transformed to the spatial domain, the gain factors are extracted and multiplied with the transformed HOA signal in the spatial domain, wherein a gain compensated transformed HOA signal is obtained. The gain compensated transformed HOA signal is transformed back into the HOA domain, wherein a gain compensated HOA signal is obtained. The DRC may be applied in the QMF-filter bank domain.

Abstract: Improved tools for authoring and rendering audio reproduction data are provided. Some such authoring tools allow audio reproduction data to be generalized for a wide variety of reproduction environments. Audio reproduction data may be authored by creating metadata for audio objects. The metadata may be created with reference to speaker zones. During the rendering process, the audio reproduction data may be reproduced according to the reproduction speaker layout of a particular reproduction environment.

Abstract: Embodiments are described for a method of rendering audio for playback through headphones comprising receiving digital audio content, receiving binaural rendering metadata generated by an authoring tool processing the received digital audio content, receiving playback metadata generated by a playback device, and combining the binaural rendering metadata and playback metadata to optimize playback of the digital audio content through the headphones.

Abstract: Embodiments are described for a high-frequency waveguide that improves the performance of large-scale surround sound and immersive audio environments. A horn waveguide is configured to be asymmetric about one of a vertical axis and horizontal axis of the waveguide to form an asymmetric horn waveguide. A spherical enclosure surrounds the asymmetric horn waveguide to form a horn speaker, and a three-axis mounting system is configured to fix the horn speaker to one of a wall or ceiling surface of the venue, wherein the mounting system facilitates rotating the horn speaker to a location that provides maximum coverage of the venue within the passband of the asymmetric horn waveguide.

Abstract: Systems, methods, and computer program products for audio processing based on convolutional neural network (CNN) are described. The CNN architecture may comprise a multi-scale input block and a multi-scale nested block. The multi-scale input block may be configured to receive input data and to generate a first downsampled input data set by downsampling the input data. The multi-scale nested block may comprise a first encoding layer configured to generate a first encoded data set by performing a convolution based on the input data. The multi-scale nested block may comprise a second encoding layer configured to generate a second encoded data set by performing a convolution based on the first downsampled input data set. Furthermore, the multi-scale nested block may comprise a first convolutional layer configured to generate a first output data set by upsampling the second encoded data set, concatenating the first encoded data set and the upsampled second encoded data set, and performing a convolution.

Abstract: In a cloud-based system for encoding high dynamic range (HDR) video, each node receives a video segment and bumper frames. Each segment is subdivided into primary scenes and secondary scenes to derive scene-based forward reshaping functions that minimize the amount of reshaping-related metadata when coding the video segment. When a parent scene of a secondary scene is processed by two or more neighboring nodes, initial forward reshaping functions and trim-pass correction parameters are adjusted using reference tone-mapping functions and updated scene-based trim-pass correction parameters.

Abstract: A projection system and calibration method therefor relate to a light source configured to emit a light in response to an image data, an illumination optical system configured to steer the light, the illumination optical system including a first lens group and a second lens group, a digital micromirror device (DMD) including a plurality of micromirrors respectively configured to reflect the steered light to a predetermined location as on-state light or to reflect the steered light as off-state light to a light dump; determining a deviation between an actual angle of orientation and an expected angle of orientation of a respective micromirror of the plurality of micromirrors; calculating a first amount of lateral adjustment corresponding to the first lens group and a second amount of lateral adjustment corresponding to the second lens group, and actuating the first and second lens groups according to the corresponding first and second amount.

Abstract: The present invention relates to a method for predicting transform coefficients representing frequency content of an adaptive block length media signal, by receiving a frame and receiving block length information indicating a number of quantized transform coefficients for each block in the frame, the number of quantized transform coefficients being one of a first or second number, wherein the first number is greater than the second number, determining a first block has the second number of quantized transform coefficients, converting the first block into a converted block having the first number of quantized transform coefficients, conditioning a main neural network trained to predict at least one output variable given at least one conditioning variable, the at least one conditioning variable being based on information regarding the converted block and block length information for the first block, providing at least one predicted transform coefficients from an output stage of the main neural network.

Abstract: Methods and apparatus for decoding a compressed Higher Order Ambisonics (HOA) representation of a sound or soundfield. The method may include receiving a bit stream containing the compressed HOA representation and decoding, based on a determination that there are multiple layers, the compressed HOA representation from the bitstream to obtain a sequence of decoded HOA representations. A first subset of the sequence of decoded HOA representations is determined based only on corresponding ambient HOA components. A second subset of the sequence of decoded HOA representations is determined based on corresponding ambient HOA components and corresponding predominant sound components.

Abstract: An audio processing unit (APU) is disclosed. The APU includes a buffer memory configured to store at least one frame of an encoded audio bitstream, where the encoded audio bitstream includes audio data and a metadata container. The metadata container includes a header and one or more metadata payloads after the header. The one or more metadata payloads include dynamic range compression (DRC) metadata, and the DRC metadata is or includes profile metadata indicative of whether the DRC metadata includes dynamic range compression (DRC) control values for use in performing dynamic range compression in accordance with at least one compression profile on audio content indicated by at least one block of the audio data.

Abstract: A system for suppressing noise and enhancing speech and a related method are disclosed. The system trains a neural network model that takes banded energies corresponding to an original noisy waveform and produces a speech value indicating the amount of speech present in each band at each frame. The neural model comprises a feature extraction block that implements some lookahead. The feature extraction block is followed by an encoder with steady down-sampling along the frequency domain forming a contracting path. The encoder is followed by a corresponding decoder with steady up-sampling along the frequency domain forming an expanding path. The decoder receives scaled output feature maps from the encoder at a corresponding level. The decoder is followed by a classification block that generates a speech value indicating an amount of speech present for each frequency band of the plurality of frequency bands at each frame of the plurality of frames.

Abstract: Using a standard-based RGB to YCbCr color transform a new RGB to YCC 3×3 transformation matrix and a 3×1 offset vector are derived under a set of coding-efficiency constraints. The new RGB to YCC 3×3 transform comprises a luminance scaling factor and a 2×2 chroma sub-matrix that preserves the energy of the standard-based RGB to YCbCr transform while maintaining or improving coding efficiency. It also adds support for an authorization or watermarking mechanism in streaming video applications. Examples of using the new color transform using image reshaping are also provided.

Abstract: A global index value is generated for selecting a global reshaping function for an input image of a relatively low dynamic range using luma codewords in the input image. Image filtering is applied to the input image to generate a filtered image. The filtered values of the filtered image provide a measure of local brightness levels in the input image. Local index values are generated for selecting specific local reshaping functions for the input image using the global index value and the filtered values of the filtered image. A reshaped image of a relatively high dynamic range is generated by reshaping the input image with the specific local reshaping functions selected using the local index values.

Abstract: Computer-implemented methods for training a neural network, as well as for implementing audio encoders and decoders via trained neural networks, are provided. The neural network may receive an input audio signal, generate an encoded audio signal and decode the encoded audio signal. A loss function generating module may receive the decoded audio signal and a ground truth audio signal, and may generate a loss function value corresponding to the decoded audio signal. Generating the loss function value may involve applying a psychoacoustic model. The neural network may be trained based on the loss function value. The training may involve updating at least one weight of the neural network.