Water detection through spatio-temporal invariant descriptors

详细信息    查看全文

• 作者：Pascal Mettes ; ^a ; ^b ; ^{P.S.M.Mettes@uva.nl} ; Robby T. Tan^a ; ^c ; ^{robbytan@unisim.edu.sg} ; Remco C. Veltkamp^a ; ^{R.C.Veltkamp@uu.nl}
• 关键词：Water detection ; Spatio-temporal descriptors ; Fourier analysis ; Invariants ; Markov random fields
• 刊名：Computer Vision and Image Understanding
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：154
• 期：Complete
• 页码：182-191
• 全文大小：2938 K
• 卷排序：154

文摘

We introduce a video pre-processing step to remove background reflections and inherent water colours. We introduce a hybrid spatial and temporal descriptor for local water classification. We introduce a new dataset, the Video Water Database, for experimental evaluation and to encourage research into water detection. We show experimentally that our water detection method improves over methods from dynamic texture and material recognition.