The THUMOS challenge on action recognition for videos “in the wild”

详细信息    查看全文

• 作者：Haroon Idrees ; ^a ; ^{haroon@cs.ucf.edu} ; Amir R. Zamir^b ; Yu-Gang Jiang^c ; Alex Gorban^e ; Ivan Laptev^d ; Rahul Sukthankar^e ; Mubarak Shah^a
• 关键词：Action recognition ; Action detection ; Action localization ; Untrimmed videos ; THUMOS ; Dataset ; Benchmark ; UCF101
• 刊名：Computer Vision and Image Understanding
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：155
• 期：Complete
• 页码：1-23
• 全文大小：8222 K
• 卷排序：155

文摘

THUMOS challenge was introduced in 2013 to serve as a benchmark for action recognition. In this paper we describe the THUMOS benchmark in detail. Give an overview of data collection and annotation procedures. Present results of submissions to the THUMOS 2015 challenge and review the participating approaches. We conclude by proposing several directions and improvements for future THUMOS challenges.