Interactive Playout of Digital Video Streams

Interactive Playout of Digital Video Streams

Samuel Rivas (LambdaStream Servicios Interactivos, Spain), Miguel Barreiro (LambdaStream Servicios Interactivos, Spain) and Víctor M. Gulías (University of A Coruña, MADS Group, Spain)
DOI: 10.4018/978-1-60566-014-1.ch101
OnDemand PDF Download:
$37.50

Abstract

Even though digital systems have many advantages over traditional analogue systems, end users expect that they will not loose any functionality in the transition. Concretely, capabilities of video cassette recorders (VCR) should be ported to stand alone digital media players (e.g., DVD players) and streaming services (e.g., digital TV and VoD services). Those capabilities are usually known as interactive playout or, simply, digital VCR functionality. Typical VCR features are random access, pause/resume, reverse play, fast forward/ backward, and slow motion. Random access and pause/resume functionalities are relatively easy to implement in digital systems. On the other hand, video coding techniques and bandwidth restrictions severely complicate the implementation of the other VCR operations. Usually, VCR capabilities apply only to video streams. In interactive playout mode, audio streams are commonly discarded. Also, the quality of video streams in interactive mode may be downgraded due to system limitations.
Chapter Preview
Top

Introduction

Interactive Playout

Even though digital systems have many advantages over traditional analogue systems, end users expect that they will not loose any functionality in the transition. Concretely, capabilities of video cassette recorders (VCR) should be ported to stand alone digital media players (e.g., DVD players) and streaming services (e.g., digital TV and VoD services). Those capabilities are usually known as interactive playout or, simply, digital VCR functionality. Typical VCR features are random access, pause/resume, reverse play, fast forward/backward, and slow motion.

Random access and pause/resume functionalities are relatively easy to implement in digital systems. On the other hand, video coding techniques and bandwidth restrictions severely complicate the implementation of the other VCR operations.

Usually, VCR capabilities apply only to video streams. In interactive playout mode, audio streams are commonly discarded. Also, the quality of video streams in interactive mode may be downgraded due to system limitations.

Architecture Constraints

Both stand-alone and client-server systems have architectures similar to Figure 1. There is a server side, where video data is stored, a data path, and a client that receives and plays the video streams. The client has a decoder and a small buffer to temporarily store data received from the server until the decoder decodes and presents them. The client buffer is very small compared to the server storage; it can only store a few seconds of video data.

Figure 1.

Multimedia service architecture

For example, in a stand-alone DVD player, the data server is the DVD reader, the storage the DVD disk, the data path the internal memory bus, and the client is the internal decoder. In a VoD service for digital TV, the server and storage are the VoD server, the data path is the digital TV network, and the client is the set top box.

There are two main differences between stand-alone and client-server systems. First, the bandwidth is much higher for data paths in stand-alone systems. Second, in client-server systems, many clients share the data server computational power.

Decoders are usually based on standards. This imposes some restrictions on the behaviour they can implement to support VCR functionality. For MPEG-2 (ISO/IEC, 1995) based decoders, an important restriction is that frame rate is fixed to a set of available values. Later MPEG standards (ISO/IEC, 1999, 2005) support arbitrary frame rates.

For the data path, the main constraint is bandwidth, especially in client-server applications.

In the server side, the main constraints are storage size and computational power.

Key Terms in this Chapter

P Picture: Picture coded with motion compensation. P pictures depend only on pictures from the past.

VoD (Video on Demand): VoD systems allow users to select and watch remote videos. VoD services contrast with traditional TV services were contents are scheduled and broadcast for all users.

Access Point: Location from where a decoder can start decoding a media stream.

Streaming: Technique used to send data streams. Instead of waiting to receive the whole data stream, streaming clients consume the input stream at the same time they receive it.

Stream: Series of bits representing real-time data.

GOP (Group of Pictures): MPEG-2 structure that contains an I picture and all the pictures that depend on it.

B Picture: Picture coded with motion compensation. B pictures may depend on one picture from the past, from one picture from the future, or both.

Motion Compensation: Video compression technique used to remove temporal redundancy between frames. Motion-compensated pictures depend on other reference pictures to be decoded.

Bitrate: Number of bits transmitted per unit of time.

Index: Data structure used to map labels and access points.

VCR (Video Cassette Recorder): Device used to record analogue TV broadcasts in magnetic video-tapes.

I Picture: Picture coded without motion compensation.

Complete Chapter List

Search this Book:
Reset