Ati Technologies Ulc

Advanced Micro Devices Corporation

Haibo Liu

Ihab Amer

Gabor Sines

Edward A Harold

Yang Liu

Wei Gao

Feng Pan

Ying Luo

Crystal Yeong-Pian Sau

Khaled Mammou

Lei Zhang

Boris Ivanovic

Andy Sung

Fabio Gulino

Ho Hin Lau

Yinan Jiang

Ahmed M Abdelkhalek

Dezhi Ming

Guopei Qiao

Zhidong Xu

Ihab M A Amer

Dong Liu

Wai Ki Lo

Jinbo Qiu

Ehsan Mirhadi

Stephen Edward Smith

Samuel D Naffziger

Greg Sadowski

Jin Li

Arun Sundaresan Iyer

Jiao Wang

Thomas E Ryan

Crystal Sau

Shu-Hsien Wu

Zhiqi Hao

Sonu Thomas

Kyle Plumadore

Sunil Gopal Koteyar

Catalin Beju

Alvin Duong

Minghao Zhu

Lawrence A Bair

Greg Sadowksi

Methods and apparatus encode image frames using intra-frame prediction by predicting pixels for a block of current pixels, based on a detected spatial pattern of pixel intensity differences among a plurality of neighboring reconstructed pixels to the block of current pixels, and encode a block of pixels of the image frame using the predicted block of reconstructed pixels. Inter-frame prediction is provided by determining whether blocks of pixels in temporally neighboring reconstructed frames corresponding to a candidate motion vector have a pattern of pixel intensity differences among the blocks from temporally neighboring frames. Predicted blocks are produced for a reconstructed frame based on the determined pattern of pixel intensity difference among temporally neighboring frames.

Video encoding/decoding using detected pattern of pixel intensity differences

Techniques for allowing the encoder to use additional data for encoding where the encoder predicts or determines that the encoder will insert filler bits into the encoded video are provided. Specifically, according to these techniques, the encoder uses a modified lambda parameter to calculate rate-distortion cost, which is then used to set encoder settings. In 'normal' operation, the encoder utilizes a lambda parameter to designate a trade-off between increasing quality (“distortion”) and reducing number of bits consumed. This trade-off is reflected in the “rate-distortion cost.” In such “normal” operation, the encoder bases this lambda value on the quantization parameter selected for a block of video to be encoded. When filler bits are predicted to be included in the encoded video data, the encoder modifies the lambda parameter to skew the trade-off more towards minimizing distortion and less towards reducing the number of bits consumed by the encoding.

Filler data-based adaptive lambda for video encoding

To leverage an amount of unused bandwidth at a hardware encoder to generate motion estimation data, a processing unit includes a hardware encoder configured to perform a first encoding job including encoder sessions to encode a captured frame, determine motion estimation data for a rendered frame, and encode the rendered frame. Further, the processing unit includes a pre-processing circuitry configured to determine a set of motion estimation parameters based on an encoder delay associated with the performance of the first encoding job by the hardware encoder. The hardware encoder is then configured to perform a second encoding job using the determined set of motion estimation parameters.

Dynamic motion estimation parameter modification for video encoding

An encoder implements a residual-free palette encoding mode in which a block of pixels is used to derive a palette table having a number of palette colors less than a number of pixel colors in the block of pixels, and to derive a color map representing each pixel of the block with a corresponding index number associated with a palette color that most closely matches the pixel's color. The calculations of residuals representing errors between the predicted palette colors and the actual pixel colors are omitted during the encoding process, thereby facilitating implementation of less complex palette mode encoder hardware at the expense of slight loss of color accuracy. Moreover, when multiple encoding modes are available, the encoder can employ the residual-free palette encoding mode when the rate-distortion cost or other cost of using this mode is determined to be the lowest cost of the plurality of encoding modes.

Residual-free palatte mode coding

A technique for varying firmware for different virtual functions in a virtualized device is provided. The virtualized device includes a hardware accelerator and a microcontroller that executes firmware. The virtualized device is virtualized in that the virtualized device performs work for different virtual functions (with different virtual functions associated with different virtual machines), each function getting a 'time-slice' during which work is performed for that function. To vary the firmware, each time the virtualized device switches from performing work for a current virtual function to work for a subsequent virtual function, one or more microcontrollers of the virtualized device examines memory storing addresses for firmware for the subsequent virtual function and begins executing the firmware for that subsequent virtual function. The addresses for the firmware are provided by a corresponding virtual machine at configuration time.

Varying firmware for virtualized device

Systems, apparatuses, and methods for performing parallel histogram calculation with application to palette table derivation are disclosed. An encoder calculates a first histogram for a first portion of pixel component value bits of a block of pixels. Then, the encoder selects a first number of the highest pixel count bins from the first histogram. Also, the encoder calculates a second histogram for a second portion of pixel component value bits of the block. The encoder selects a second number of the highest pixel count bins from the second histogram. A third histogram is calculated from the concatenation of bits assigned to the first and second number of bins, and the highest pixel count bins are selected from the third histogram. A palette table is derived based on these highest pixel count bins selected from the third histogram, and the block of pixels is encoded using the palette table.

Parallel histogram calculation with application to palette table derivation

Systems, apparatuses, and methods for calculating multi-pass histograms for palette table derivation are disclosed. An encoder calculates a first histogram for a first portion of most significant bits (MSBs) of pixel component values of a block of an image or video frame. Then, the encoder selects a given number of the highest pixel count bins from the first histogram. The encoder then increases the granularity of these selected highest pixel count bins by evaluating one or more additional bits from the pixel component values. A second histogram is calculated for the concatenation of the original first portion MSBs from the highest pixel count bins and the one or more additional bits, and the highest pixel count bins are selected from the second histogram. A palette table is derived based on these highest pixel count bins selected from the second histogram, and the block is encoded using the palette table.

Hierarchical histogram calculation with application to palette table derivation

Methods and devices are provided for encoding a video stream which comprise encoding a plurality of frames of video acquired from different points of view, generating statistical values for the frames of video determined from values of pixels of the frames, generating, for each of the plurality of frames, a perceptual hash value based on statistical values of the frame and encoding a current frame comprising video acquired from a corresponding one of the different points of view using a previously encoded reference frame based on a similarity of perceptual hashes of the current frame and the previously encoded reference frame.

Identifying long term reference frame using scene detection and perceptual hashing

Systems, apparatuses, and methods for generating a model for determining a quantization strength to use when encoding video frames are disclosed. A pre-encoder performs multiple encoding passes using different quantization strengths on a portion or the entirety of one or more pre-processed video frames. The pre-encoder captures the bit-size of the encoded output for each of the multiple encoding passes. Then, based on the multiple encoding passes, the pre-encoder generates a model for mapping bit-size to quantization strength for encoding video frames or portion(s) thereof. When the encoder begins the final encoding pass for one or more given video frames or any portion(s) thereof, the encoder uses the model to map a preferred bit-size to a given quantization strength. The encoder uses the given quantization strength when encoding the given video frame(s) or frame portion(s) to meet a specified bit-rate for the encoded bitstream.

Content adaptive quantization strength and bitrate modeling

A system and method for dynamically changing encode quality at a block level based on runtime pre-encoding analysis of content in a video stream. A video encoder continuously analyzes the content during runtime, and collects statistics and/or characteristics of the content before encoding it. This classifies the block among pre-defined categories of content, where every category has its own compression parameters.

Hierarchical measurement of spatial activity for text/edge detection

Systems, apparatuses, and methods for calculating multi-pass histograms for palette table derivation include an encoder that calculates a first histogram for a first portion of most significant bits (MSBs) of pixel component values of a block of an image or video frame. Then, the encoder selects a given number of the highest pixel count bins from the first histogram. The encoder then increases the granularity of these selected highest pixel count bins by evaluating one or more additional bits from the pixel component values. A second histogram is calculated for the concatenation of the original first portion MSBs from the highest pixel count bins and the one or more additional bits, and the highest pixel count bins are selected from the second histogram. A palette table is derived based on these highest pixel count bins selected from the second histogram, and the block is encoded using the palette table.

Various codecs and methods of using the same are disclosed. In one aspect, a method of processing video data is provided that includes encoding or decoding the video data with a codec in aggressive deployment and correcting one or more errors in the encoding or decoding wherein the error correction includes re-encoding or re-decoding the video data in a non-aggressive deployment or generating a skip picture.

Adaptive error-controlled dynamic voltage and frequency scaling for low power video codecs

Growing community of inventors

Markham, Canada

Haibo Liu