A frame, in a video context, is a single still image that, when played in sequence with the other frames of the video, creates motion on the playback surface. Analog frames are physical cels of film brought across a projector light at high speed. Digital frames are encoded by software implementing a codec. These frames may be a complete image or may be a transformation of other frames in the video sequence; clever removal of duplicated data and expressing it in terms of those transformations is the basis of much of digital video compression.
Major video codecs often use the following categories of frames to express a video:
- an intra-coded picture, or I-frame, is a complete image, akin to an inline PNG or JPG.
- a predictive coded picture, or P-frame, expresses only the changes relative to previous frames--which in some encoders may not have to be the immediately preceding frame.
- a bipredictive coded picture, or B-frame, expresses changes from both previous and future frames.
Frames can also be further divided into slices in many major codecs.