Netflix Inc.

Anush Moorthy

Zhi Li

Christos G Bampis

Anne Aaron

Megha Manohara

Aditya Mavlankar

Liwei Guo

Li-Heng Chen

Francisco J San Miguel

Naveen Mareddy

Ameya Ramesh Vasani

Guanhua Ye

Xiaomei Liu

Chih Hao Lin

Dmitry Vasilyev

Adithya Prakash

Kyle Christopher Swanson

In various embodiments, a video editing application receives, from an encoding application, a list of random access points of an encoded media content item. When an operator specifies a point of interest for the media content item, the video editing application determines a closest random access point to the point of interest and modifies the point of interest to coincide with the closest random access point. The encoding application can also modify the point of interest to coincide with an existing random access point if the point of interest is within a tolerance of the existing random access point and the point of interest is not frame specific. Further, the encoding application can modify media metrics, computed using pre-encoded versions of the media content item, using heuristics to account for operator-specified points of interest that do not coincide with, or are not within a tolerance of, existing random access points.

Techniques for avoiding re-encoding of media content items

In various embodiments, a serverless function agent determines that a client stub function has been invoked with a first set of arguments in a first execution environment. The serverless function agent then performs one or more operations on a media item that is associated with a first argument included in the first set of arguments to generate a second argument included in a second set of arguments. Notably, the first argument has a first data type and the second argument has a second data type. Subsequently, the serverless function agent invokes a function with the second set of arguments in a second execution environment. Advantageously, because the serverless function agent automatically performs operations on the media item, the overall amount of technical know-how and manual effort required to enable the function to successfully execute on a wide range of media items can be reduced.

Techniques for executing serverless functions on media items

In various embodiments, a perceptual quality application determines an absolute quality score for encoded video content viewed on a target viewing device. In operation, the perceptual quality application determines a baseline absolute quality score for the encoded video content viewed on a baseline viewing device. Subsequently, the perceptual quality application determines that a target value for a type of the target viewing device does not match a base value for the type of the baseline viewing device. The perceptual quality application computes an absolute quality score for the encoded video content viewed on the target viewing device based on the baseline absolute quality score and the target value. Because the absolute quality score is independent of the viewing device, the absolute quality score accurately reflects the perceived quality of a wide range of encoded video content when decoded and viewed on a viewing device.

Device-consistent techniques for predicting absolute perceptual video quality

In various embodiments, an encoding ladder application generates encoding ladders for encoding media titles. In operation, the encoding ladder application generates a first convex hull representing encoding tradeoffs between quality and bitrate when encoding a media title at a first resolution; The encoding ladder application generates a second convex hull representing encoding tradeoffs between quality and bitrate when encoding the media title at a second resolution. Based on the first convex hull and the second convex hull, the encoding ladder application generates an overall convex hull. Subsequently, the encoding ladder application generates an encoding ladder for the media title based on at least the overall convex hull and a ladder requirement. Advantageously, the tradeoffs between quality and bitrate represented by the encoding ladder are customized for the media title. Consequently, encoding inefficiencies attributable to conventional fixed-bitrate ladders can be reduced.

Techniques for generating per-title encoding ladders

In various embodiments, a function build application compiles source code to generate an executable version of a function that has a first function signature. The function build application then replaces a first data type of a first parameter included in the first function signature with a second data type to generate a second function signature for a client stub function. Subsequently, the function build application generates a remote procedure call (RPC) client that includes the client stub function. Notably, the RPC client causes the function to execute when the client stub function is invoked. Advantageously, unlike conventional techniques that require manual generation of strongly typed functions, the function build application automatically customizes the RPC client for the function.

Techniques for interfacing between media processing workflows and serverless functions

In various embodiments, a training application trains a machine learning model to preprocess images. In operation, the training application computes a chroma sampling factor based on a downscaling factor and a chroma subsampling ratio. The training application executes a machine learning model that is associated with the chroma sampling factor on data that corresponds to both an image and a first chroma component to generate preprocessed data corresponding to the first chroma component. Based on the preprocessed data, the training application updates at least one parameter of the machine learning model to generate a trained machine learning model that is associated with the first chroma component.

Machine learning techniques for component-based image preprocessing

In various embodiments, a shot collation application causes multiple encoding instances to encode a source video sequence that includes at least two shot sequences. The shot collation application assigns a first shot sequence to a first chunk. Subsequently, the shot collation application determines that a second shot sequence does not meet a collation criterion with respect to the first chunk. Consequently, the shot collation application assigns the second shot sequence or a third shot sequence derived from the second shot sequence to a second chunk. The shot collation application causes a first encoding instance to independently encode each shot sequence assigned to the first chunk. Similarly, the shot collation application causes a second encoding instance to independently encode each shot sequence assigned to the second chunk. Finally, a chunk assembler combines the first encoded chunk and the second encoded chunk to generate an encoded video sequence.

Techniques for optimizing encoding tasks

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.

Growing community of inventors

Redwood City, CA, United States of America

Anush Moorthy