首页 实验室简介 研究方向 研究队伍 项目成果 学术会议 课程建设 人才招聘 留言板 英文站
 

标 题: Video Affective Content Representation and Recognition Based on Video Affective Tree and HMMs
作 者: Kai Sun, Junqing Yu
会议 / 期刊: Proceeding of the 6th International Conference on Entertainment Computing, 249~2
发 表 时 间: 2007年
下 载 地 址: 点击下载
论文摘要
Video affective content analysis is a fascinating but seldom addressed field in entertainment computing research communiti-
es. To recognize affective content in video, a video affective content representation and recognition framework based on Video Af
fective Tree (VAT) and Hidden Markov Models (HMMs) was proposed. The proposed video affective contentrecognizer has good potentia
l to recognize the basic emotional events ofaudience. However, due to Expectation-Maximization (EM) methods like the Baum-Welch a
lgorithm tend to converge to the local optimum which is the closer to the starting values of the optimization procedure, the esti
mation of the recognizer parameters requires a more careful examination. A Genetic Algorithm combined HMM (GA-HMM) is presented h
ere to address this problem. The idea is to combine a genetic algorithm to explore quickly the whole solution space with a Baum-W
elch algorithm to find the exact parameter values of the optimum. The experimental results show that GA-HMM can achieve higher re
cognition rate with less computation compared with our previous works.

Copyright @2007 华中科技大学数字媒体处理与检索实验室 版权所有
地址:华中科技大学南一楼西部441 电话:13871273335