Chapter 9:  Memory Management:Codecs
by
Bhaskaran Vasudev and Wei Li
presented by
Jong Chih Chien


I. Introduction 

    Advanced in recent years in memory management areas include robust algorithms for reducing size of image, video, or audio data and vlsi architectures for relatively low-cost but sophisticated compression processes, thus spurring developments of codecs. [ref: Multimedia Telecommunication News]  A codec is a system that will encode and decode multimedia information; generally as compressor and decompressor.
 
  This implies that we may be able to treat multimedia information as just another data type that can be manipulated, transformed, stored, and transmitted as complete information; not as individual types cobbled together.  This presents the possibility of standardization across different platforms and applications.  Different standards for compression methodologies will be presented from algorithmic and architectural viewpoints for image and video data will be presented.
 
II. Background

   The basic task of a codec is to compact a given signal for efficient representation.  Table 1 presents a list for applications that can benefit for a standardized codec.
 
 

Application
Data
Uncompressed
Rate 
Compressed
Voice 
8k samples/s, 8 bits/sample
64 kbps
2-4 kbps
Slow-motion video (10 fps) 
framesize 176x120, 8 bits/pixel
5.07 Mbps
8-16 kbps
Audio conference 
8k samples/s, 8 bits/sample
64 kbps
16-64 kbps
Video conference (15 fps) 
framesize 352x240, 8 bits/pixel
30.41 Mbps
64-768 kbps
Digital audio (stereo) 
44.1k samples/s, 16bits/sample
1.5 Mbps
0.128-1.5 Mbps
Video file transfer (15 fps) 
framesize 352x240, 8 bits/pixel
30.41 Mbps
384 kbps
Digital video on CD-ROM (30 fps) 
framesize 352x240, 8 bits/pixel
60.83 Mbps
1.5-4 Mbps
Broadcast video (30 fps) 
framesize 720x480, 8 bits/pixel
248.83 Mbps
3-8 Mbps
HDTV (59.94 fps) 
framesize 1280 x 720, 8 bits/pixel
1.33 Gbps
20 Mbps
Table 1  Applications for image, video and audio compression.