Temporal Action Proposal (TAP) generation is an important problem, as fast and accurate extraction of semantically important (e.g. human actions) segments from untrimmed videos is an important step for large-scale video analysis. We propose a novel Temporal Unit Regression Network (TURN) model. There are two salient aspects of TURN: (1) TURN jointly predicts action proposals and refines the temporal boundaries by temporal coordinate regression; (2) Fast computation is enabled by unit feature reuse: a long untrimmed video is decomposed into video units, which are reused as basic building blocks of temporal proposals. TURN outperforms the previous state-of-theart methods under average recall (AR) by a large margin on THUMOS-14 and ActivityNet datasets, and runs at over 880 frames per second (FPS) on a TITAN X GPU. We further apply TURN as a proposal generation stage for existing temporal action localization pipelines, it outperforms stateof-the-art performance on THUMOS-14 and ActivityNet.
In the simple color-magnetic interaction model, we investigate possible ground cscs tetraquark states in the diquark-antidiquark basis. We use several methods to estimate the mass spectrum and discuss possible assignment for the X states observed in the J/ψφ channel. We find that assigning the Belle X(4350) as a 0 ++ tetraquark is consistent with the tetraquark interpretation for the X(4140) and X(4270) while the interpretation of the X(4500) and X(4700) needs orbital or radial excitation. There probably exist several tetraquarks around 4.3 GeV that decay into J/ψφ or ηcφ.
In this work, we study systematically the mass splittings of the qqQQ (q = u, d, s and Q = c, b) tetraquark states with the color-magnetic interaction by considering color mixing effects and estimate roughly their masses. We find that the color mixing effect is relatively important for the J P = 0 + states and possible stable tetraquarks exist in the nnQQ (n = u, d) and nsQQ systems either with J = 0 or with J = 1. Possible decay patterns of the tetraquarks are briefly discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.