Where should cameras look at soccer games: Improving smoothness using the overlapped hidden Markov model.

Computer Vision and Image Understanding(2017)

引用 24|浏览40
暂无评分
摘要
•A novel Overlapped Hidden Markov Model (OHMM) is proposed to smoothly predict camera angles for soccer game broadcasting.•A highly automatic camera calibration method (CalibMe) is proposed for soccer video calibration.•The OHMM and CalibMe are extensively evaluated on synthetic and real data. OHMM outperforms the state-of-the-art algorithms in terms of smoothness, and CalibMe outperforms strong baselines in terms of accuracy.
更多
查看译文
关键词
Camera planning,Camera calibration,Hidden Markov model,Soccer games
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要