High Efficiency Video Coding (HEVC) is the latest buzzword in the video compression. HEVC envisions to attain the same visual quality as H.264/MPEG-4 Advanced Video Coding (AVC) with just 50 % of the bit rate requirement of current video standard (H.264/MPEG-4). Two to ten time increase in computational complexity is permitted. HVEC relies on VLSI and parallel computing to achieve the 'bit rate reduction' feat. There is no paradigm shift in the way video compression is carried out. Instead HEVC employs computationally intensive but compression efficient algorithms. Reference [1] is suited for enthusiasts and [2] for researchers.
In an MPEG video encoder, video is divided into frames, slices, macroblock and block. Block is the smallest unit that contains 8x8 pixels. MPEG exploit temporal (literally time; frame-to-frame) and spatial redundancies (like JPEG). Because of this very high compression ratio is achieved. To improve compression, MPEG uses a concept of motion vector. It employs Discrete Cosine Transform (DCT). Redundancy removed video is encoded by Huffman coding (popular variable length coding technique) to form bit stream. Importantly MPEG specify the format of compressed video only. Encoding the video and implementation of hardware is left to the manufacturers.
