公开(公告)号：US20170295374A1

公开(公告)日：2017-10-12

申请号：US15207468

申请日：2016-07-11

Applicant: NETFLIX, Inc.

Inventor： Anne AARON ,  Zhi LI ,  Todd GOODALL

IPC: H04N19/154 ,  H04N17/00 ,  G06K9/00

Abstract: In various embodiments, a quality trainer trains a model that computes a value for a perceptual video quality metric for encoded video content. During a pre-training phase, the quality trainer partitions baseline values for metrics that describe baseline encoded video content into partitions based on genre. The quality trainer then performs cross-validation operations on the partitions to optimize hyperparameters associated with the model. Subsequently, during a training phase, the quality trainer performs training operations on the model that includes the optimized hyperparameters based on the baseline values for the metrics to generate a trained model. The trained model accurately tracks the video quality for the baseline encoded video content. Further, because the cross-validation operations minimize any potential overfitting, the trained model accurately and consistently predicts perceived video quality for non-baseline encoded video content across a wide range of genres.

公开(公告)号：US20240121402A1

公开(公告)日：2024-04-11

申请号：US17937033

申请日：2022-09-30

Applicant: NETFLIX, INC.

Inventor： Christos G. BAMPIS ,  Zhi LI

IPC: H04N19/154 ,  H04N19/136 ,  H04N19/172 ,  H04N19/182 ,  H04N19/184 ,  H04N19/42

CPC classification number: H04N19/154 ,  H04N19/136 ,  H04N19/172 ,  H04N19/182 ,  H04N19/184 ,  H04N19/42

Abstract: In various embodiments, a quality inference application estimates perceived video quality for reconstructed video. The quality inference application computes a set of feature values corresponding to a set of visual quality metrics based on a reconstructed frame, a source frame, a display resolution, and a normalized viewing distance. The quality inference application executes a trained perceptual quality model on the set of feature values to generate a perceptual quality score that indicates a perceived visual quality level for the reconstructed frame. The quality inference application performs one or more operations associated with an encoding process based on the perceptual quality score.

公开(公告)号：US20230144735A1

公开(公告)日：2023-05-11

申请号：US17981281

申请日：2022-11-04

Applicant: NETFLIX, INC.

Inventor： Christos G. BAMPIS ,  Zhi LI

IPC: G06T3/40

CPC classification number: G06T3/4046

Abstract: In various embodiments a training application trains convolutional neural networks (CNNs) to reduce reconstruction errors. The training application executes a first CNN on a source image having a first resolution to generate a downscaled image having a second resolution. The training application executes a second CNN on the downscaled image to generate a reconstructed image having the first resolution. The training application computes a reconstruction error based on the reconstructed image and the source image. The training application updates a first learnable parameter value included in the first CNN based on the reconstruction error to generate at least a partially trained downscaling CNN. The training application updates a second learnable parameter included in the second CNN based on the reconstruction error to generate at least a partially trained upscaling CNN.

公开(公告)号：US20220217429A1

公开(公告)日：2022-07-07

申请号：US17700231

申请日：2022-03-21

Applicant: NETFLIX, INC.

Inventor： Christos BAMPIS ,  Zhi LI

IPC: H04N21/24 ,  H04N21/25

Abstract: In various embodiments, a quality of experience (QoE) prediction application computes a visual quality score associated with a stream of encoded video content. The QoE prediction application also determines a rebuffering duration associated with the stream of encoded video content. Subsequently, the QoE prediction application computes an overall QoE score associated with the stream of encoded video content based on the visual quality score, the rebuffering duration, and an exponential QoE model. The exponential QoE model is generated using a plurality of subjective QoE scores and a linear regression model. The overall QoE score indicates a quality level of a user experience when viewing reconstructed video content derived from the stream of encoded video content.

公开(公告)号：US20210127120A1

公开(公告)日：2021-04-29

申请号：US17141081

申请日：2021-01-04

Applicant: NETFLIX, INC.

Inventor： Zhi LI ,  Christos BAMPIS

IPC: H04N19/154 ,  H04N19/91 ,  H04N21/235 ,  G06N3/08 ,  G06K9/62 ,  G06N3/04 ,  G06N20/20 ,  G06N20/10 ,  G06N5/00 ,  H04N17/00

Abstract: In various embodiments, a prediction application computes a quality score for re-constructed visual content that is derived from visual content. The prediction application generates a frame difference matrix based on two frames included in the re-constructed video content. The prediction application then generates a first entropy matrix based on the frame difference matrix and a first scale. Subsequently, the prediction application computes a first value for a first temporal feature based on the first entropy matrix and a second entropy matrix associated with both the visual content and the first scale. The prediction application computes a quality score for the re-constructed video content based on the first value, a second value for a second temporal feature associated with a second scale, and a machine learning model that is trained using subjective quality scores. The quality score indicates a level of visual quality associated with streamed video content.

公开(公告)号：US20190246111A1

公开(公告)日：2019-08-08

申请号：US15890709

申请日：2018-02-07

Applicant: NETFLIX, INC.

Inventor： Zhi LI ,  Christos BAMPIS

IPC: H04N19/154 ,  H04N19/91 ,  G06K9/62 ,  G06N3/08 ,  H04N21/235

CPC classification number: H04N19/154 ,  G06K9/6269 ,  G06N3/08 ,  H04N17/004 ,  H04N19/91 ,  H04N21/235

Abstract: In various embodiments, a prediction application computes a quality score for re-constructed visual content that is derived from visual content. The prediction application generates a frame difference matrix based on two frames included in the re-constructed video content. The prediction application then generates a first entropy matrix based on the frame difference matrix and a first scale. Subsequently, the prediction application computes a first value for a first temporal feature based on the first entropy matrix and a second entropy matrix associated with both the visual content and the first scale. The prediction application computes a quality score for the re-constructed video content based on the first value, a second value for a second temporal feature associated with a second scale, and a machine learning model that is trained using subjective quality scores. The quality score indicates a level of visual quality associated with streamed video content.

公开(公告)号：US20180167619A1

公开(公告)日：2018-06-14

申请号：US15782586

申请日：2017-10-12

Applicant: NETFLIX, INC.

Inventor： Zhi LI ,  Anne AARON ,  Anush MOORTHY ,  Christos BAMPIS

IPC: H04N19/154 ,  H04N19/59 ,  H04N19/146 ,  H04N19/593

Abstract: In various embodiments, a perceptual quality application computes an absolute quality score for encoded video content. In operation, the perceptual quality application selects a model based on the spatial resolution of the video content from which the encoded video content is derived. The model associates a set of objective values for a set of objective quality metrics with an absolute quality score. The perceptual quality application determines a set of target objective values for the objective quality metrics based on the encoded video content. Subsequently, the perceptual quality application computes the absolute quality score for the encoded video content based on the selected model and the set of target objective values. Because the absolute quality score is independent of the quality of the video content, the absolute quality score accurately reflects the perceived quality of a wide range of encoded video content when decoded and viewed.

公开(公告)号：US20220103869A1

公开(公告)日：2022-03-31

申请号：US17549793

申请日：2021-12-13

Applicant: NETFLIX, INC.

Inventor： Zhi LI

IPC: H04N19/86

Abstract: In various embodiments, a tunable VMAF application reduces an amount of influence that image enhancement operations have on perceptual video quality estimates. In operation, the tunable VMAF application computes a first value for a first visual quality metric based on reconstructed video content and a first enhancement gain limit. The tunable VMAF application computes a second value for a second visual quality metric based on the reconstructed video content and a second enhancement gain limit. Subsequently, the tunable VMAF application generates a feature value vector based on the first value for the first visual quality metric and the second value for the second visual quality metric. The tunable VMAF application executes a VMAF model based on the feature value vector to generate a tuned VMAF score that accounts, at least in part, for at least one image enhancement operation used to generate the reconstructed video content.

公开(公告)号：US20220038710A1

公开(公告)日：2022-02-03

申请号：US16945737

申请日：2020-07-31

Applicant: NETFLIX, INC.

Inventor： Zhi LI

IPC: H04N19/149 ,  H04N19/154 ,  H04L29/06

Abstract: In various embodiments, a data optimization application mitigates scoring inaccuracies in subjective quality experiments. In operation, the data optimization application generates a model that includes a first set of individual scores and a first set of parameters. The first set of parameters includes a first subjective score set and a first set of subjective factor sets. The data optimization application performs one or more optimization operations on the first set of parameters to generate a second set of parameters. The second set of parameters includes a second subjective score set and a second set of subjective factor sets, wherein the second subjective score set compensates for at least a first subjective factor set included in the second set of subjective factor sets. The data optimization application also computes a participant evaluation report based on at least a second subjective factor sets included in the second set of subjective factor sets

公开(公告)号：US20210092178A1

公开(公告)日：2021-03-25

申请号：US17113884

申请日：2020-12-07

Applicant: NETFLIX, INC.

Inventor： Te-Yuan HUANG ,  Chaitanya EKANADHAM ,  Andrew J. BERGLUND ,  Zhi LI

IPC: H04L29/06 ,  H04L12/26

Abstract: In various embodiments, a hindsight application computes a total download size for a sequence of encoded chunks associated with a media title for evaluation of at least one aspect of a video streaming service. The hindsight application computes a feasible download end time associated with a source chunk of the media title based on a network throughput trace and a subsequent feasible download end time associated with a subsequent source chunk of the media title. The hindsight application then selects an encoded chunk associated with the source chunk based on the network throughput trace, the feasible download end time, and a preceding download end time associated with a preceding source chunk of the media title. Subsequently, the hindsight application computes the total download size based on the number of encoded bits included in the first encoded chunk. The total download size correlates to an upper bound on visual quality.