Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Unified Buffer Approach for Compressed and Uncompressed Video Data on Multimedia Adapters

IP.com Disclosure Number: IPCOM000117279D
Original Publication Date: 1996-Jan-01
Included in the Prior Art Database: 2005-Mar-31
Document File: 2 page(s) / 84K

Publishing Venue

IBM

Related People

Cook, RS: AUTHOR [+3]

Abstract

Video capture/compression cards typically need separate memory buffers to hold the uncompressed and compressed data. Fig. 1 shows a traditional approach to a video capture/compression adapter card. The live video is input to the video processor as well as to the video memory. In capture applications, the video processor scales the image size, converts it to RGB ( Red, Green and Blue) color space, dithers the data and stores it in its frame buffer to be sent to display for monitoring. At the same time, the Video Codec compresses the data in the video memory. The Codec requires an output buffer for the compressed data due to the burst nature of the compression algorithm. During playback, the compressed data is read into the buffer from the system memory or a LAN adapter and decompressed into video memory by the Codec.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Unified Buffer Approach for Compressed and Uncompressed Video Data
on Multimedia Adapters

      Video capture/compression cards typically need separate memory
buffers to hold the uncompressed and compressed data.  Fig. 1 shows a
traditional approach to a video capture/compression adapter card.
The live video is input to the  video processor as well as to the
video memory.  In capture applications, the video processor scales
the image size, converts it to RGB ( Red, Green and Blue) color
space, dithers the data and stores it in its frame buffer to be sent
to display for monitoring.  At the same time, the Video Codec
compresses the data in the video memory.  The Codec requires an
output buffer for the compressed data due to the burst nature of the
compression algorithm.  During playback, the compressed data is read
into the buffer from the system memory or a LAN adapter and
decompressed into video memory by the Codec.  The video processor
performs color space conversion, scaling, dithering on the data and
stores in video processor's frame buffer to be sent for display.
Typically a FIFO (First-in First-out) buffer is used for compressed
data that has a very high cost associated with it.  Having many
buffers also adds cost and complexity to both the hardware and
software design.

      Fig. 2 illustrates the unified buffer approach on Ultimedia
Video I/O and JPEG (Joint Photographic Experts Group) standard
adapter for both the compressed and uncompressed video data.  The
compressed data buffer is completely eliminated and instead the video
processor's frame buffer is used.  By making use of the host
interface to both the video processor (PX1070) and Codec (CL550), the
compressed data can be shuffled in both directions.  During the
capture operation, input video in the video memory is compressed by
the Codec and put in a portion of the...