Video Compression Basics
A video is a sequence of images being shown. In video these images are called frames and the amount of images shown per second is called frames per second. The more frames per second a video has, the more smooth and realistic it will appear when shown.
The aim of compression is to reduce the file size. The backside is that to reduce file size you reduce information, which often translates to reduced quality. However, not all data, contained in a raw footage, are equally important. The main challenge for compression software is to identify what information can safely be removed or - preferably - be replaced by less space demanding, but sufficient information to recreate the original video at playback time.
Compression software utilizes a number of more or less advanced techniques. One technique may be the better for some parts of the video another technique may be the better for other parts. Often a number of different techniques are applied to the same parts. Some techniques are lossless - i.e. the video can be fully restored to its original shape. Other techniques are lossy. The art of the lossy techniques is to confine the loss to what is undetectable by the human eye. Different programs differ in how well they are in particularly this part.
Lossless Compression Techniques
An example of a lossless compression technique works by removing the parts of each subsequent frame that were already shown in a previous frame. For example, imagine a video clip of a talking person, who is standing still in a room where nothing moves except his mouth. On the first frame the image is shown complete, but on the subsequent frames only the area around the mouth will be shown as long as nothing else changes.
It follows that the size of a video showing an hour long monologue can be reduced considerably by this technique alone while for an action movie, where the action is going on all over the whole screen, more sophisticated techniques have to be applied.
Lossy Compression Techniques
An example of a lossy technique is one that compresses the individual frames with a method similar to the one used when compressing jpeg images. Just like you can't recover the full quality image from the lower quality jpeg, you can't restore the quality of the original video at playback time. Nevertheless, just like a compressed jpeg may suffice, if its only meant to be shown on a screen, so may a video compressed by this method suffice and even improve the user's overall experience because of the shorter download time.
