As described above, compression techniques are used to eliminate the spatial, temporal and statistical redundancy of the video signal. Generally, they can be categorised in lossless and lossy video coding schemes. The former are preserving every bit of the original information, producing identical reconstructed pictures at the decoder, while the latter sacrifice some of the least significant visual information to improve compression efficiency. Currently, lossless schemes are only used in medical imaging applications, where, even for psychological reasons, it is justifiable to maintain maximal decoding quality at the expense of performance, while lossy schemes have been adopted for nearly all other applications.
The algorithms are usually applied in two stages; first waveform encoding and then entropy encoding. Some of the most popular generic algorithms for coding of video sequences, i.e. content-independent schemes, are described in the following paragraphs.