Paper Info
Title
Unsupervised random forest indexing for fast action search
Abstract
Despite recent successes of searching small object in images, it remains a challenging problem to search and locate actions in crowded videos because of (1) the large variations of human actions and (2) the intensive computational cost of searching the video space. To address these challenges, we propose a fast action search and localization method that supports relevance feedback from the user. By characterizing videos as spatio-temporal interest points and building a random forest to index and match these points, our query matching is robust and efficient. To enable efficient action localization, we propose a coarse-to-fine sub-volume search scheme, which is several orders faster than the existing video branch and bound search. The challenging cross-dataset search of several actions validates the effectiveness and efficiency of our method.
Year
DOI
Venue
2011
10.1109/CVPR.2011.5995488
CVPR
Keywords
Field
DocType
localization method,video signal processing,query matching,spatio temporal interest points,coarse-to-fine sub-volume search scheme,existing video branch,challenging cross-dataset search,crowded video,video space searching,crowded videos,indexing,human actions,video branch-and-bound search,fast action search,challenging problem,object detection,object searching,relevance feedback,efficient action localization,unsupervised random forest indexing,coarse-to-fine subvolume search scheme,human action,bound search,query processing,neodymium,vegetation
Computer vision,Object detection,Branch and bound,Relevance feedback,Pattern recognition,Information retrieval,Computer science,Search engine indexing,Beam search,Artificial intelligence,Random forest
Conference
Volume
Issue
ISSN
2011
1
1063-6919
ISBN
Citations 
PageRank 
978-1-4577-0394-2
41
1.26
References 
Authors
27
3
Name
Order
Citations
PageRank
Gang Yu138219.85
Junsong Yuan23703187.68
zicheng liu33662199.64