Which frames are utilized for bi-predictive construction of an image?

Bi-predictive construction of an image in video coding refers to the use of frames that can predict their content from both preceding and succeeding frames. B-frames, or Bi-directional frames, are specifically designed for this purpose. They achieve compression by using information from both the preceding frame (I-frame or P-frame) and the following frame to reconstruct the image.

This method allows B-frames to leverage temporal data more effectively than I-frames and P-frames, which only use one reference for their predictive calculations. As a result, B-frames can provide better compression and quality since they can utilize more context from the surrounding frames during the encoding process. This distinguishing feature of B-frames in bi-predictive construction makes them essential in various video coding standards, such as H.264 and HEVC, which aim to optimize storage and transmission of video data.

Utilizing B-frames thus enhances the overall efficiency of the video compression process, making them critical for achieving high-quality video in limited bandwidth scenarios.