LEAD MPEG-2 Encoder (3.0)

The LEAD MPEG-2 Encoder Filter is a DirectShow® filter for compressing video data using the MPEG-2 (ISO/IEC 13818-2) or MPEG-1 (ISO 11172-2) standard. The LEAD MPEG-2 Decoder is a DirectShow® filter for decompressing MPEG-2 video data.

MPEG-1 is the older standard and achieves higher compression than MPEG-2 (but at lower quality). MPEG-1 compression is used in VCD (video on CD).

MPEG-2 is the newer standard and has lower compression than MPEG-1 but the quality is much better than MPEG-1. MPEG-2 is a high quality video compression algorithm and is best suited for applications with higher bit rates or high-bandwidth usage. The most common applications include DVD-Videos, SVCD and digital TVs. The compressed video data can be stored inside MPEG, AVI , DICOM or Ogg files with the option of saving the file with or without the audio data.

The DICOM files can store MPEG-2 SD or HD video. See "Supplement 42: MPEG2 Transfer Syntax" and "Supplement 137: MPEG2 MP@HL Transfer Syntax" of the DICOM specification for more information on requirements of the MPEG-2 SD or HD video.

The compression options for the LEAD MPEG-2 Encoder can be set manually through the encoder property page. The compression options can also be set programmatically using the ILMMPG2Encoder interface.

The maximum video size for MPEG2 video is 1920x1152 (when using the High profile).

Filter Interfaces	ILMMPG2Encoder, ILMMCommon
Interface ID	IID_ILMMPG2Encoder
Input Media Types	Type: MEDIATYPE_Video Subtypes: MEDIASUBTYPE_YV12 MEDIASUBTYPE_YUY2 MEDIASUBTYPE_UYVY MEDIASUBTYPE_RGB32 MEDIASUBTYPE_RGB24
Output Media Types	Type: MEDIATYPE_Video Subtype: MEDIASUBTYPE_MPEG2_VIDEO MEDIASUBTYPE_LMPG2
Filter CLSID	CLSID_LMMPEG2Encoder
Property Page CLSID	CLSID_LMMPEG2EncoderProperty
Executable	LENCMPG23.dll, DSKernel2.dll
Filter category	CLSID_VideoCompressorCategory
Header	ILEncMPEG2_3.h

Platforms

Win32, x64

Levels and Profiles

A MPEG-2 encoder capable of handling all resolutions, bit rates and frame rates up to HDTV would be too expensive for many simpler applications or hardware implementations. Therefore, the MPEG-2 standard was subdivided into levels and profiles. The MPEG-2 standard allows for four source formats, or Levels, to be coded; ranging from Limited Definition (todays VCR quality), to full HDTV, each with a range of bit rates. In addition to this flexibility in source formats, MPEG-2 allows for different Profiles. Each profile offers a collection of compression tools that together make up the coding system. A different profile means that a different set of compression tools is available. The four levels, including frame size and maximum bit rate, are defined in the following table:

Level	Maximum Frame size (PAL/NTSC	Maximum Bitrate	Significance
Low	352x288 352x240	4 Mb/s	CIF, consumer tape equiv.
Main	720x576 720x480	15 Mb/s	ITU-R 601, Studio TV
High 1440	1440x1152 1440x1080	60 Mb/s	4x 601, consumer HDTV
High	1920x1152 1920x1080	80 Mb/s	PROD. SMPTE

Most of today's work is done at the Main Level, which is designated for standard video. Within this level, the Main profile is used for many applications. The MPEG-2 encoder will automatically adjust the level if the video being encoded is bigger than the maximum size for the current level.

I Frames and P Frames

To increase the compression performance, MPEG-2 uses Temporal Compression. Temporal compression is achieved by encoding only the difference between successive frames instead of the frames themselves. This means that frames cannot be considered in isolation. A given picture is constructed from the prediction from a previous picture, and may be used to predict the next picture. An MPEG-2 video stream is comprised of I Frames, B Frames and P Frames:

I: (Intraframe) frames. These are encoded without reference to another frame to allow for random access.
P: (Predictive) frames are encoded using the previous I (or P) frame as reference. P frames are compressed more than I frames.
B: (Bi-directional) frames are encoded using both previous and next I (or P) frames as reference and they are compressed more than both I, and P frames.

To increase the compression of MPEG-2, more B and P frames should be used. You do so by adjusting the "I Frame" and "P Frame" options on the Encoder property page:

I Frame: represents the distance between I frames. All the frames between successive I frames are P or B frames.
P Frame: represents the distance between P frames. All the frames between successive frames are B frames.

For example if I frame is 12, and P Frame is 3, then video stream looks like the following:

I B B P B B P B B P B B I B B P B B P B B P B B ..

The distance between I frames must be a multiple of the distance between the P frames. The LEAD MPEG-2 Encoder will automatically adjust these values to preserve this relationship.

This filter is included as part of the following modules:

LEAD MPEG-2 Video Codec Module