Video affective content recognition based on genetic algorithm combined HMM

Video affective content analysis is a fascinating but seldom addressed field in entertainment computing research communities. To recognize affective content in video, a video affective content representation and recognition framework based on Video Affective Tree (VAT) and Hidden Markov Models (HMMs) was proposed. The proposed video affective content recognizer has good potential to recognize the basic emotional events of audience. However, due to Expectation-Maximization (EM) methods like the Baum-Welch algorithm tend to converge to the local optimum which is the closer to the starting values of the optimization procedure, the estimation of the recognizer parameters requires a more careful examination. A Genetic Algorithm combined HMM (GA-HMM) is presented here to address this problem. The idea is to combine a genetic algorithm to explore quickly the whole solution space with a Baum-Welch algorithm to find the exact parameter values of the optimum. The experimental results show that GA-HMM can achieve higher recognition rate with less computation compared with our previous works.