For The Love of Computing: Matching Your Eye to Images and Video — DCT and Quantization

So when does a 90TB one minute video become a 1MB file?

We take so many things for granted in this modern age. On our cameras, we create images using hi-res cameras, which we can easily upload. But, uncompressed, the files for a 20 Megapixel camera would be 50MB each, and if we recorded video at 30 frames per second (30 fps) it would be 90,000,000,000 Bytes (90TB) for every minute of video. So how do our cameras, DVD players and digital TV manage to cram so much information into such a small amount of data?

The magic of DCT and Quantization

The magic all happens underneath, and where the compression system knows exactly what your eyes will see. Imagine the bandwidth that would be required for watching a football match on TV if it wasn’t for compression — and especially the DCT (Discrete Cosine Transform) method? One minute of our football match would eat up the bandwidth our the whole of your area.

Remember too that video is basically a whole lot of images being played back … a bit like a flick book that you played with as a child.

Your wonderful eyes

Your eyes are wonderful things and can process and make sense of complex things so much…