Qualcomm Incorporated

Amir Said

Marta Karczewicz

Vadim Seregin

Xin Zhao

Hilmi Enes Egilmez

Jianle Chen

Taco Sebastiaan Cohen

Reza Pourreza

Yinhao Zhu

Wei-Jung Chien

Yang Yang

Li Feng Zhang

Vijayaraghavan Thirumalai

Cheng-Teh Hsieh

Hoang Cong Minh Le

Nan Hu

Yung-Hsuan Chao

Jie Dong

Yizhe Zhang

Yadong Lu

Kai Zhang

Xiang Yang Li

Rajan Laxman Joshi

Geert Van Der Auwera

Akshay Gadde

Auke Joris Wiggers

Reza Pourreza Shahri

Dana Kianfar

Systems and techniques are provided for coding (e.g., encoding and/or decoding) video data using camera motion information. For example, a decoding device can obtain a frame of encoded video data associated with an input frame, the frame of encoded video data including camera information associated with generating the video data and a residual. A camera motion compensated frame can be generated based on a reference frame and the camera information. Optical flow information associated with object motion determined based on at least the input frame and the reference frame can be generated. A motion compensated frame can be generated by warping the camera motion compensated frame based on the optical flow information. A reconstructed input frame can be generated based on the motion compensated frame and the residual.

Video coding using camera motion compensation and object motion compensation

This disclosure describes entropy coding techniques for media data coded using neural-based techniques. A media coder is configured to determine a probability distribution function parameter for a data element of a data stream coded by a neural-based media compression technique, wherein the probability distribution function parameter is a logarithmic function of a standard deviation of a probability distribution function of the data stream, determine a code vector based on the probability distribution function parameter, and entropy code the data element using the code vector.

Entropy coding for neural-based media compression

A device may perform a first prediction process for a first block of video data to produce a first residual. The device may apply a first transform process to the first residual to generate first transform coefficients for the first block of video data and encode the first transform coefficients. The device may perform a second prediction process for a second block of video data to produce a second residual. The device may determine that a second transform process, which includes the first transform process and at least one of a pre-adjustment operation or a post-adjustment operation, is to be applied to the second residual. The device may apply the first transform process and the pre- or post-adjustment operation to the second residual to generate second transform coefficients for the second block. The coding device may code the first and second transform coefficients.

Multiple transforms adjustment stages for video coding

A processor-implemented method for image compression using an artificial neural network (ANN) includes receiving, at an encoder of the ANN, an image and a spatial segmentation map corresponding to the image. The spatial segmentation map indicates one or more regions of interest. The encoder compresses the image according to a controllable spatial bit allocation. The controllable spatial bit allocation is based on a learned quantization bin size.

Neural image compression with controllable spatial bit allocation

A video encoder determines scaled transform coefficients, wherein determining the scaled transform coefficients comprises scaling transform coefficients of a block of the video data according to a given quantization step. The video encoder determines scalar quantized coefficients, wherein determining the scalar quantized coefficients comprises applying scalar quantization to the scaled transform coefficients of the block. Additionally, the video encoder applies a neural network that determines a respective set of probabilities for each respective transform coefficient of the block. The respective set of probabilities for the respective transform coefficient includes a respective probability value for each possible adjustment value in a plurality of possible adjustment values. Inputs to the neural network include the scaled transform coefficients and the scalar quantized coefficients. The video encoder determines, based on the set of probabilities for a particular transform coefficient of the block, a quantization level for the particular transform coefficient.

Parallelized rate-distortion optimized quantization using deep learning

Certain aspects of the present disclosure provide techniques for compressing content using a neural network. An example method generally includes receiving content for compression. The content is encoded into a first latent code space through an encoder implemented by an artificial neural network trained to generate a latent space representation of the content. A first compressed version of the encoded content is generated using a first quantization bin size of a series of quantization bin sizes. A refined compressed version of the encoded content is generated by scaling the first compressed version of the encoded content into one or more second quantization bin sizes smaller than the first quantization bin size, conditioned at least on a value of the first compressed version of the encoded content. The refined compressed version of the encoded content is output for transmission.

Progressive data compression using artificial neural networks

A computer-implemented method for operating an artificial neural network (ANN) includes receiving an input by the ANN. The ANN generates a latent representation of the input. The latent representation is communicated according to a bit rate based on a learned latent scaling parameter. The latent scaling parameter is learned based on a channel index and a tradeoff parameter value that corresponds to a value that balances the bit rate and a distortion.

Variable bit rate compression using neural network models

This disclosure describes entropy coding techniques for media data coded using neural-based techniques. A media coder is configured to determine a probability distribution function parameter for a data element of a data stream coded by a neural-based media compression technique, wherein the probability distribution function parameter is a function of a standard deviation of a probability distribution function of the data stream, determine a code vector based on the probability distribution function parameter, and entropy code the data element using the code vector.

Neural-network media compression using quantized entropy coding distribution parameters

Techniques are described herein for processing video data. For instance, a technique can include encoding one or more frames of video data using a video encoder, the video encoder including at least a quantization process; determining an actual bit rate of the encoded one or more frames; predicting an estimated bit rate using an encoder proxy, the encoder proxy including a statistical model for estimating a bit rate of the encoded one or more frames; determining, using the encoder proxy, a gradient of the estimated bit rate; and training the encoder proxy to predict the estimated bit rate based on the actual bit rate, the estimated bit rate, and the gradient.

Bit-rate estimation for video coding with machine learning enhancement

A method of image compression includes receiving an image. Multiple quantized latent representations are generated to represent features of the image. Each of the quantized latent representations has a different resolution and is generated at staggered timings. Each of the later generated quantized latent representations is conditioned on each of the prior generated quantized latent representations. The multiple quantized latent representations are decoded to reconstruct the image.

Data compression with a multi-scale autoencoder

An example method of decoding video data includes obtaining, from a coded video bitstream and for a current block of the video data, an indication of an intra-prediction mode that identifies an initial predictive block; filtering, in parallel, samples in a current line of a plurality of lines of the initial predictive block based on filtered values of samples in a preceding line of the plurality of lines and unfiltered values of samples in the current line to generate filtered values for samples for the current line; and reconstructing, using intra prediction, values of samples of the current block based on the filtered values of the samples of the current initial predictive block and residual data for the current block that represents a difference between the filtered values of the samples of the current initial predictive block and the values of samples of the current block.

Video intra prediction using hybrid recursive filters

Techniques are described herein for processing video data. For instance, a process can include obtaining encoded video data. The process can include determining an intersection of values between values for a first termination byte of a first parcel of the encoded video data and values of a second termination byte of a second parcel of the encoded video data. The process can further include determining a joint termination byte for the first termination byte of the first parcel and the second termination byte of the second parcel. Values for the joint termination byte are based on the intersection of values. The process can include generating entropy coded data including the joint termination byte for the first parcel and the second parcel. The entropy coded data can be generated using arithmetic coding or binary coding.

Joint termination of bidirectional data blocks for parallel coding

Systems and techniques are described for encoding and/or decoding data based on motion estimation that applies variable-scale warping. An encoding device can receive an input frame and a reference frame that depict a scene at different times. The encoding device can generate an optical flow identifying movements in the scene between the two frames. The encoding device can generate a weight map identifying how finely or coarsely the reference frame can be warped for input frame prediction. The encoding device can generate encoded video data based on the optical flow and the weight map. A decoding device can generate a reconstructed optical flow and a reconstructed weight map from the encoded data. A decoding device can generate a prediction frame by warping the reference frame based on the reconstructed optical flow and the reconstructed weight map. The decoding device can generate a reconstructed input frame based on the prediction frame.

Multi-scale optical flow for learned video compression

Techniques are described of using Position Dependent Intra Prediction Combination (PDPC). A video coder such as a video encoder or a video decoder utilizes PDPC in cases where a current block intra mode predicted using an angular intra prediction mode.

Growing community of inventors

San Diego, CA, United States of America

Amir Said