Surveillance Video Coding with Background Preprocessing is used to filter the background from the current video frame. Typically in video surveillance the background image is constant and not a point of interest. The foreground imagery carries the important surveillance information. Therefore, the foreground is encoded separately with a minimum of compression artifacts.
Video coding standards H.264/FVC and H.265/HEVC provide efficient video encoding and decoding. However, these standards are usually not the optimal choices for video surveillance since they are designed for general video applications. Efficient compression and storage of surveillance video is essential.
To do efficient foreground compression, the background has to be eliminated from the current video frame and only the residual video that is the foreground is encoded. In many cases this is only a small part of the current frame. The background is extracted from the video sequence by the background filter and encoded as a separate video stream. The background is updated much slower than the foreground.
For video reconstruction, the background and foreground sequences must be decoded and combined to be displayed.