Video Stream Encoding with Sequence Header Index for Adaptive Switching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional adaptive video streaming techniques experience playback interruptions due to time-consuming frame matching processes when switching between video streams, disrupting the user experience.

Innovation Solution

A method for encoding video streams that involves applying a video codec to generate groups of pictures (GOPs) with specific playback time intervals and offsets, using the Advanced Systems Format (ASF) to create data packets, and generating a sequence header index with switch points for efficient switching between encoded video streams of different bit rates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the content player sequentially searches the new video stream to locate the matching video frame when switching between video streams, then the video frame matching can be achieved, but the searching operation is very time consuming and causes playback interruption

Engineering Contradiction:
Improvevideo frame matching accuracyVSAvoidsearching time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-generating and storing an index of key frame locations and timestamps during the video encoding phase. This index allows the content player to directly locate matching frames without sequential searching, thus resolving the contradiction between achieving accurate frame matching and minimizing search time.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If the content player downloads a different content file with higher playback bit rate when bandwidth increases, then playback quality is maximized, but the switching between different bit rate streams causes playback interruption

Engineering Contradiction:
Improveplayback qualityVSAvoidswitching interruption time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent prepares switching information in advance during encoding, including indexes of key frames and their timestamps across multiple bit rate streams. This preliminary preparation enables the content player to perform rapid, interruption-free switching between different quality streams by directly navigating to the appropriate pre-identified frame locations.

Inventive Principle:
Principle #10Preliminary action

3Stability of the object's composition

If the content player sequentially searches for matching video frames when switching between video streams, then frame synchronization is maintained, but the downloading process is disrupted

Engineering Contradiction:
Improveframe synchronizationVSAvoiddownloading efficiency
Core Design Contradiction:
Stability of the object's compositionVSProductivity

Solution Approach 1:

The patent pre-computes and stores the temporal relationships and positions of key frames in an index structure during the encoding phase. This preliminary action maintains frame synchronization accuracy while eliminating the need for time-consuming sequential searches during playback, thus preserving downloading efficiency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an index structure as an intermediary between the video stream data and the content player's frame matching process. This index acts as a lookup table that provides direct access to frame locations, maintaining synchronization without disrupting the downloading process.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP2467956B1Encoding video streams for adaptive video streaming
Publication Date: 2015.09.30 NETFLIX INC
  • EP2467956B1 patent drawingFigure 1
  • EP2467956B1 patent drawingFigure 2
  • EP2467956B1 patent drawingFigure 3

AI summary

One embodiment of the invention sets forth an encoding server including components configured to encode a video stream associated with a content title for adaptive streaming. The video stream is first processed by a VC1 encoder to generate an encoded video stream comprising a multiple GOPs, each GOP including a key frame and having a different playback offset. The encoded video stream is then packaged such that the GOPs are stored in data packets of the packaged encoded stream. An SHI generator generates an SHI associated with the packaged encoded stream that includes a switch point associated with each GOP. Each switch point includes the playback offset associated with the corresponding GOP and the data packet storing the key frame of the corresponding GOP. The SHI associated with multiple packaged encoded video streams associated with the same content title and encoded to different playback bit rates have corresponding switch points.