MPEG Standards: MPEG-1, MPEG-2, MPEG-4

• MPEG stands for Moving Picture Expert Group.
• MPEG worked to generate the specifications under ISO and IEC.
• "MPEG video" presently consists of two finalized standards, MPEG-1 and MPEG-2.
• MPEG-4 was finalized in 1998 for Very Low Bitrate Audio-Visual Coding.
• The MPEG-1 and MPEG-2 standards are similar in basic concepts like motion compensated block-based transform coding techniques.
• MPEG-4 deviates from these by using software image construct descriptors for target bit-rates in the very low range, < 64Kb/sec.
• Focus is on MPEG-1 and MPEG-2 compression techniques due to their widespread use.
• There is no MPEG-3 standard because minor extensions to MPEG-2 sufficed for HDTV applications.
• MPEG-7 "Multimedia Content Description Interface" completion was scheduled for July 2001.
• MPEG-21 "Multimedia Framework" work started in June 2000, producing a Draft Technical Report and two Calls for Proposals.
• MPEG-1 was finalized in 1991 and optimized for video resolutions of 352x240 pixels at 30 frames/sec (NTSC based) or 352x288 pixels at 25 frames/sec (PAL based), referred to as Source Input Format (SIF) video.
• MPEG-1 resolution may go as high as 4095x4095 at 60 frames/sec.
• The bit-rate is optimized for applications of around 1.5 Mb/sec.
• MPEG-1 is defined for progressive frames only, without support for interlaced video applications.
• MPEG-2 was finalized in 1994, addressing digital television broadcasting issues like field-interlaced video coding and scalability.
• The target bit-rate for MPEG-2 was raised to between 4 and 9 Mb/sec, potentially resulting in very high quality video.
• MPEG-2 consists of profiles defining bitstream scalability and colorspace resolution, and levels defining image resolution and maximum bit-rate per profile.
• Main Profile, Main Level (MP@ML) refers to 720x480 resolution video at 30 frames/sec, at bit-rates up to 15 Mb/sec for NTSC video.
• Main Profile, High Level (MP@HL) descriptor can achieve a HDTV resolution of 1920x1080 pixels at 30 frame/sec, at a bit-rate of up to 80 Mb/sec.
• MPEG compression attempts to overcome some shortcomings of H.261 and JPEG.

Basic Steps Used in Video Compression

• In standards such as MPEG 1, 2 and H.263.
  • Motion Estimation.
  • Motion Compensation and Image Subtraction.
  • Discrete Cosine Transform.
  • Quantization.
  • Run Length Encoding.
  • Entropy Coding – Huffman Coding.
• The MPEG solution is to add a third frame type (B-frame) which is a bidirectional frame.
• B-frames search for macroblocks in past and future frames.
• Typical frame patterns: IBBPBBPBB, etc.

MPEG Video Layers

• MPEG video is organized into a hierarchy of layers to aid in error handling, random search/editing, and synchronization with audio.
• The top level is the video sequence layer any self-contained bitstream like a movie or advertisement.
• The second layer is the group of pictures, containing one or more groups of intra (I) frames and/or non-intra (P and/or B) pictures.
• The third layer is the picture layer itself, followed by the slice layer.
• Each slice is a contiguous sequence of raster ordered macroblocks.
• Each slice consists of macroblocks, which are 16x16 arrays of luminance pixels, or picture data elements, plus 2 8x8 arrays of associated chrominance pixels.
• Macroblocks can be divided into distinct 8x8 blocks for further processing, such as transform coding.
• Each layer has a unique 32 bit start code, typically with 23 zero bits followed by a one, then 8 bits for the actual start code.

B-Frames

• MPEG encoders can use forward/backward interpolated prediction, these frames are referred to as bi-directional interpolated prediction frames, or B frames for short.
• In sequence I,B,P,B,P,B,I,B,P,B,P,B, I frames are coded spatially, P frames are forward predicted, and B frames are based using forward and backward prediction from I or P frames.
• Backward prediction requires encoding and transmitting future frames first, out of order.
• There is no limit to the number of consecutive B frames that could be used.
• Broadcast applications use 2 consecutive B frames (I,B,B,P,B,B,P,).

B-Frame Encoding

• The main advantage of using B frames is improved coding efficiency.
• Using these frames means fewer bits are coded.
• Quality can also improve with moving objects revealing otherwise hidden areas.
• Backward prediction enables smarter encoding decisions.
• Since B frames do not predict future frames, errors do not propagate.
• One disadvantage is the frame reconstruction memory buffers within the encoder and decoder must be doubled in size to accommodate the 2 anchor frames.
• Another disadvantage is the delay throughout the system due to out-of-order frame delivery.