Budget Amount *help |
¥8,580,000 (Direct Cost: ¥6,600,000、Indirect Cost: ¥1,980,000)
Fiscal Year 2015: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000)
Fiscal Year 2014: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
Fiscal Year 2013: ¥3,640,000 (Direct Cost: ¥2,800,000、Indirect Cost: ¥840,000)
|
Outline of Final Research Achievements |
Most facial expression recognition models eventually create a classifier based on expression images taken during a short period of time, and use them as the base data for learning. However, because so many facial expression patterns exist, representations cannot be made of all of them, and it is difficult to obtain all available patterns within a short period of time and retain them for use as learning data. For a facial expression recognition model to retain its high robustness continuously along the time axis over a long period of time, the classifier created during the initial stage should evolve and gradually become adaptive over time. In other words, it is necessary for the model to retain its existing knowledge and simultaneously learn by adding newly available knowledge as it becomes available. We propose a method for creating a facial expression recognition model that can offer such an adaptive learning capability.
|