“Fundamentals of Video Compression Techniques” by Daniel

  • ©Andy Daniel

Conference:


Entry Number: 09

Title:

    Fundamentals of Video Compression Techniques

Course Organizer(s):



Presenter(s):



Abstract:


    Prerequisites
    A basic understanding of bit-mapped graphics. Understanding of the concept of the pixel as an RGB triplet, the image as a rectangular array of pixels, and the video clip as a sequence of images. No serious math skill required.

    Topics Covered
    The basics of digital color science, subsampling, quantization, rounding, error diffusion, delta coding, color spaces, spatial correlation, prediction, discrete cosine transforms, Huffman coding, macroblocks, and decompression.

    Description
    A picture may be worth 1,000 words, but it takes far more than 1,000 words to display a high-resolution image. When motion video is considered, the storage requirements become gigantic. Fortunately, techniques exist to lessen this data requirement without affecting the resulting image quality too badly. This course investigated and compared various compression techniques in still and motion video. It explored the underlying techniques that form the foundation of today’s compression algorithms, with an emphasis on examining the trade-offs between quantity and quality for different techniques.


Contributed By:


    Mary Whitton

Location:


    Charles Babbage Institute Archives, University of Minnesota

Type: