Patent Application
20080094414
In the following figures, same numerals are provided for the reference of same elements in the following preferred embodiment to make the illustration of a multimedia video generation system in accordance with the preferred embodiment of the present invention easier to understand.
Referring to
The characteristic recognition unit 11 is provided for recognizing a characteristic 16 in the frames 15 to obtain an attribute parameter of the characteristic 16, such as detecting a face image characteristic or a face expression image characteristic in the frame, and the attribute parameter includes a position, a size or a rotation angle of the characteristic. The characteristic recognition unit 11 carries out a characteristic recognition and a characteristic matching to obtain the position of the characteristic and carries out a tracking, wherein the characteristic recognition may consider capturing a low-level characteristic (such as feature points) or a high-level characteristic (including a face characteristic such as an eye, a mouth or a nose) based on the nature of application. The method for matching a characteristic includes an implicit algorithm and an explicit algorithm, and the explicit characteristic matching method searches a one-to-one correspondence among the characteristics, and the implicit characteristic matching method uses a parameter or a transformation to represent the relation between the characteristics of two successive frames. With the foregoing technological combination, characteristics can be detected according to different natures. For example, a combination of explicit algorithm and high-level characteristic can be used for analyzing face expression, and a combination of implicit algorithm and high-level characteristic can be used for recognizing and positioning a sense organ of a face. The characteristic recognition technology is a prior art, and thus will not be described here.
The object providing unit 12 is for providing a first object 121 and a second object 122 according to the video 14 and the characteristic 16. The displaying position of the first object corresponds to the frame 15, and the displaying position of the second object 122 corresponds to the position of the characteristic 16. The object providing unit 12 can provide a first object and a second object according to a pre-selected mode, if needed. The objects are selected from a medium material which includes a pattern, an image or an audio, and the pre-selected mode could be a festival theme such as New Year, Christmas, Mid-Autumn Festival or a cartoon character such as Superman, Spiderman, King of Monkey or a monster. Each theme includes a media material corresponding to the first object and a medium material corresponding to the second object. If the pre-selected mode is Mid-Autumn Festival, then the medium material corresponding to the first object could be a pattern of moon and cloud displayable around the frame of the video 14, and the second object could be a pattern of Moon Goddess's hair ornament displayable on a human face in the frame and moved together with the face to change the position, size or rotation angle of the display.
The video synthesis unit 13 synthesizes the video 14 with the first object 121 or the second object 122 to generate a synthesized video 17. Referring to
Preferably, the multimedia video generation system utilizes a processor to execute a program code by software.
Referring to
Step 30: executing an application program, wherein the application program provides a user interface;
Step 31: opening a video file to obtain a plurality of consecutive frames, and displaying the frames through the user interface;
Step 32: setting a synthesis theme through the user interface;
Step 33: loading a medium material corresponding to the synthesis theme, and decoding the medium material, wherein the medium material includes a first pattern and a second pattern;
Step 34: recognizing a face characteristic in the plurality of frames and carrying out a tracking to obtain an attribute parameter such as a position, a size and a rotation angle of a face characteristic in every frame;
Step 35: adjusting a second pattern according to the attribute parameter; and
Step 36: synthesizing the frames, first pattern and adjusted second pattern to generate a synthesized video file.
If the video file is an encoded video file when Step 31 is carried out, then the encoded video file will obtain a plurality of consecutive frames by the decoding step. In addition, Step 31 further includes selecting a desired processing frame through the user interface, so that users need not to wait for editing after the synthesized video file is generated.
Before Step 36 is carried out, the method further includes a preview of the synthesis result. Since the synthesized video file requires more computations and longer computing time, the preview function lets users view the synthesis ahead of time to determine whether or not the synthesis result can meet a user's expectation; if yes, then carry out Step 36, or else return to Step 32.
While the invention has been described by way of example and in terms of a preferred embodiment, it is to be understood that the invention is not limited thereto. To the contrary, it is intended to cover various modifications and similar arrangements and procedures, and the scope of the appended claims therefore should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements and procedures.
| Number | Date | Country | Kind |
|---|---|---|---|
| 095218615 | Oct 2006 | TW | national |
