This disclosure describes techniques for resizing multimedia content for efficient statistical multiplexing. In response to a request to resize a current segment of data, an encoding module associated with the selected segment adjusts the amount of motion information to be encoded to resize the segment of data. For example, the encoding module associated with the selected segment of data may merge two or more motion vectors to reduce the amount of motion information to be encoded. As another example, the encoding module reselects encoding modes for one or more blocks of pixels of at least one frame within the segment of data.