摘要 |
<p>PURPOSE:To input data with the action of the mouth of a speaker by inputting picture data concerning the mouth of the speaker and processing a picture based on the prescribed discrimination reference concerning the speaking contents. CONSTITUTION:An input picture from a camera 1 to photograph the action of the lips of a speaker is processed by a picture processing part 2 and sent to a picture analyzing part 3. For a registering processing part 4, the analyzing result of the picture is first registered and processed for the input from a special speaker and the processing result is stored in a registering part 5. Thereafter, from a collation recognizing part 6 comparing the analyzing data from the picture analyzing part 3 with the registering data, input data based on the picture to photograph the action of the inputted lips can be obtained. Thus, the data can be inputted by the action of the mouth of the speaker.</p> |