Computer Software Resources

The image in a single video frame is divided into blocks of 16x16 pixels, and each of them is divided in subsequent sub-blocks of 8x8.

12 Nov

How does video compression work

Posted in Computers on 12.11.11

The image in a single video frame is divided into blocks of 16x16 pixels, and each of them is divided in subsequent sub-blocks of 8x8. The IDCT algorithm (Inverse Discrete Cosine Transform - discrete cosine transform), used by avs video converter software, tries to merge the texture and color of pixels in a single 8x8 block and then apply the same procedure to the macro block. Depending on the level of compression applied, this merging can result in a high, medium or relatively poor image quality. In areas of solid color, for example a clear road with the background of sky, MPEG1 is effective. Its compression is low with minimal loss of detail. However, in scenes rich in detail, such as the vision of the interior of a forest, you can sense certain distorted images around objects and artifacts due to high compression.

In MPEG there are three types of frames, which are I-frames (Intra frame), the key frames of reference not compressed, the P-frames (predictive frames), the more compressed frames, and finally the B-frames (Bidirectional Predictive frame) whose structure is derived from the previous frames. The avs4you software includes a conversion tool which can convert among any video format. Read this avs video converter review for more details about this tool.