Automatic Analysis of Facial Expressions: The State of the Art
IEEE Transactions on Pattern Analysis and Machine Intelligence
Visual affect recognition
| Hi-index | 0.00 |
Two kinds of discrete type Hopield neural networks are used to recognize the facial expressions. It is expected that the associative memory effect of the Hop-field neural networks is capable to classify the preliminarily defined specific expressions among the variety of facial expressions. In this paper, 4 kinds of facial expressions such as "surprise", "anger", "happiness" and "sadness" expressed by 10 persons are used as the input data to recognize. The each image of the faces under test are divided into 8 x 10 regions as the feature area. Accordingly , 8 x 10 ternary values [+1 (moving upward), 0 (no movement), -1 (moving downward)] computed from averaged value of the optical flows in each region are used as the feature parameters. Two kinds of neural networks learnt by different learning data are cascade-connected to compensate each other. As the experimental result, the averaged recognition rate for those 4 expressions was obtained 92.2 %.