3. Generation of Multimedia Thumbnails

The proposed Multimedia Thumbnail generation algorithm accepts a scanned documents as well as an electronic document as an input and is comprised of the following steps as depicted in Figure 2:

1. Semantic Analysis and Extraction: The proposed Multimedia Thumbnail generation algorithm accepts a scanned documents as well as an electronic document as an input and is comprised of the following steps as depicted in Figure 2:
2. MMNail Optimization: A subset of extracted document elements is selected to be included in the Multimedia Thumbnail. Information attributes that indicate the importance of each document element are determined based on user and task constraints. Time attributes, that determine the time required to present each document element, are computed based on display constraints. Then, given a time duration constraint, those attributes are used to optimize the selection of document elements to be included in the MMNail.
3. Composition into a playable form and MMNail representation: Selected Multimedia Thumbnail elements, animation instructions or rendered animations, and synthesized speech are represented in a file with or without the original document file.

Each of these steps is explained in more detail in the next subsections.

Figure 2. Overview of MMNail generation steps.