Accurately estimating rigid transformations in registration using a boosting-inspired mechanism
详细信息 查看全文
- 作者：Yonghuai Liua ; yyl@aber.ac.uk" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Honghai Liub ; honghai.liu@port.ac.uk" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Ralph R. Martinc ; ralph@cs.cf.ac.uk" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Luigi De Dominicisd ; luigi.dedominicis@enea.it" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Ran Songe ; R.Song@brighton.ac.uk" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Yitian Zhaof ; Yitian.Zhao@bit.edu.cn" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Feature extraction ; Feature matching ; Point match evaluation ; Boosting-inspired ; Rigid underlying transformation
- 刊名：Pattern Recognition
- 出版年：2016
- 出版时间：December 2016
- 年：2016
- 卷：60
- 期：Complete
- 页码：849-862
- 全文大小：7257 K
文摘
- •
We provide a novel method for evaluating point matches established by typical methods.
- •
Real data based experimental results show that the method outperforms the state-of-the-art.
- •
Under certain conditions, the finally estimated underlying transformation is optimal.
- •
The transformation can be recovered with an error as small as 5% from typical point matches.