1,438 Hits in 4.5 sec

Learning Hierarchical Models of Complex Daily Activities from Annotated Videos

Jawad Tayyub, Majd Hawasly, David C. Hogg, Anthony G. Cohn
2018 2018 IEEE Winter Conference on Applications of Computer Vision (WACV)  
In this paper, we propose a novel approach for building models of complex long-term activities.  ...  This variability allows for extracting the inherent hierarchical structure of the activity in a video.  ...  Coverage defines a ratio of the number of input symbols described by the selected parse tree.  ... 
doi:10.1109/wacv.2018.00182 dblp:conf/wacv/TayyubHHC18 fatcat:n4ylaqbv7vdlljfkiryp4c6ef4

Coarse-to-Fine Classification of Road Infrastructure Elements from Mobile Point Clouds Using Symmetric Ensemble Point Network and Euclidean Cluster Extraction

Wang, Wang, Scaioni, Si
2019 Sensors  
The overall accuracy of the SEP-ECE method improves by 3.97% with respect to PointNet.  ...  The ECE subsequently adjusts points that have been predicted incorrectly by the first step.  ...  Acknowledgments: The Stanford 3D semantic parsing data sets and Mostrule Co., LTD are acknowledged for providing the public and experimental data sets, respectively.  ... 
doi:10.3390/s20010225 pmid:31906105 pmcid:PMC6982872 fatcat:i3xqmooyzbaptdcsmezo6ywkji

DendroBLAST: Approximate Phylogenetic Trees in the Absence of Multiple Sequence Alignments

Steven Kelly, Philip K. Maini, Marc Robinson-Rechavi
2013 PLoS ONE  
This method differs from other methods by incorporating a simple model of sequence evolution to test the effect of introducing sequence changes on the reliability of the bipartitions in the inferred tree  ...  Using realistic simulated sequence data we demonstrate that this method produces phylogenetic trees that are more accurate than other commonly-used distance based methods though not as accurate as maximum  ...  Acknowledgments SK would also like to thank Keith Gull, Chris Norbury and Robert Scotland for their helpful comments on the manuscript.  ... 
doi:10.1371/journal.pone.0058537 pmid:23554899 pmcid:PMC3598851 fatcat:izr2ce6n7bewjgjxiwlgsp4nha

Human gene correlation analysis (HGCA): A tool for the identification of transcriptionally co-expressed genes

Ioannis Michalopoulos, Georgios A Pavlopoulos, Apostolos Malatras, Alexandros Karelas, Myrto-Areti Kostadima, Reinhard Schneider, Sophia Kossida
2012 BMC Research Notes  
gene cluster.  ...  correlation matrix using the Neighbour Joining (NJ) clustering method.  ...  Acknowledgements Research supported by: •HMOD Offset Benefits •PEP Attica •KUL PFV/10/016 SymBioSys  ... 
doi:10.1186/1756-0500-5-265 pmid:22672625 pmcid:PMC3441226 fatcat:d5r5w27yffhujehf6j5ydfle5m

Image classification using compression distance [article]

Yuxuan Lan, Richard Harvey
2005 International Conference on Vision, Video and Graphics  
The normalised compression distance measures the mutual compressibility of two signals. We show that this distance can be used for classification on real images.  ...  The new classifier operating on these trees produces results that are very similar to those obtained on the raw images thus allowing, for the first time, classification using the full trees.  ...  Then our compression-based measurement can be applied for pair of tree strings in terms of similarity distance of these trees.  ... 
doi:10.2312/vvg.20051023 dblp:conf/vvg/LanH05 fatcat:uvmvz2ckwzejlevglltp6hnk34

Joint RNN-Based Greedy Parsing and Word Composition [article]

Joël Legrand, Ronan Collobert
2015 arXiv   pre-print
The composition procedure outputs a vector representation which summarizes syntactically (parsing tags) and semantically (words) sub-trees.  ...  This paper introduces a greedy parser based on neural networks, which leverages a new compositional sub-tree representation.  ...  ACKNOWLEDGMENTS Part of this work was supported by NEC Laboratories America.  ... 
arXiv:1412.7028v4 fatcat:ifaeb2t7evgr3fpkd6cg3qtl74

The Second Multilingual Surface Realisation Shared Task (SR'19): Overview and Evaluation Results

Simon Mille, Anja Belz, Bernd Bohnet, Yvette Graham, Leo Wanner
2019 Proceedings of the 2nd Workshop on Multilingual Surface Realisation (MSR 2019)  
For full descriptions of the participating systems, please see the separate system reports elsewhere in this volume.  ...  Systems were evaluated (a) automatically, using a range of intrinsic metrics, and (b) by human judges in terms of readability and meaning similarity to a reference.  ...  Acknowledgments SR'19 is endorsed by SIGGEN. The work on its organisation, realisation, and evaluation was supported in part by (1)  ... 
doi:10.18653/v1/d19-6301 dblp:conf/emnlp/MilleBBGW19 fatcat:fitut6g7ynbbncselbi3tqv3qi

Generative Incremental Dependency Parsing with Neural Networks

Jan Buys, Phil Blunsom
2015 Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)  
We propose a neural network model for scalable generative transition-based dependency parsing.  ...  A probability distribution over both sentences and transition sequences is parameterised by a feedforward neural network.  ...  The vocabulary is clustered into approximately p |V | classes using Brown clustering (Brown et al., 1992) , reducing the number of items to sum over in the normalisation factor from O(|V |) to O( p |V  ... 
doi:10.3115/v1/p15-2142 dblp:conf/acl/BuysB15 fatcat:u4m2j2dywzaxhpn3s7tk467oxe

Neural Syntactic Generative Models with Exact Marginalization

Jan Buys, Phil Blunsom
2018 Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)  
For supervised dependency parsing, our model achieves a stateof-the-art result among generative approaches.  ...  Our algorithms complement previous approaches by supporting batched training and enabling online computation of next word probabilities.  ...  Acknowledgments We thank members of the Oxford NLP group for discussions, Yejin Choi for valuable feedback, and the anonymous reviewers for their comments.  ... 
doi:10.18653/v1/n18-1086 dblp:conf/naacl/BuysB18 fatcat:xktysnlgijftpaxuedgd4wxz3a

Geometry Aware Mappings for High Dimensional Sparse Factors [article]

Avradeep Bhowmik, Nathan Liu, Erheng Zhong, Badri Narayan Bhaskar, Suju Rajan
2016 arXiv   pre-print
We develop techniques that use geometry aware permutation maps on a tessellated unit sphere to obtain high dimensional sparse embeddings for latent factors with sparsity patterns related to angular closeness  ...  The parse-tree procedure can be extended to the case of a D-ary tessellation scheme in a simple manner by considering parse trees where each non-leaf node has D child nodes.  ...  Many other parse-tree methods are possible.  ... 
arXiv:1605.04764v1 fatcat:eyatvuixj5gdxniavqlzlo4tp4

Genome analysis following a national increase in Scarlet Fever in England 2014

Victoria Chalker, Aleksey Jironkin, Juliana Coelho, Ali Al-Shahib, Steve Platt, Georgia Kapatai, Roger Daniel, Chenchal Dhami, Marisa Laranjeira, Timothy Chambers, Rebecca Guy, Theresa Lamagni (+4 others)
2017 BMC Genomics  
Prior to data normalisation, emm3 was the predominant type (accounting for 42.9% of SF isolates, 130/303 95%CI 37.5-48.5; 14.7% higher than the percentage of emm3 isolates found in the iGAS isolates).  ...  Contemporaneous iGAS isolates for which genomic data were available were included for comparison.  ...  The iGAS data used for comparison is collated routinely by the Respiratory and Systemic Bacteria Section, PHE as part of the national surveillance for iGAS in England and Wales.  ... 
doi:10.1186/s12864-017-3603-z pmid:28283023 pmcid:PMC5345146 fatcat:bt4as7t2mvdz7i36fz73utfqee

Explaining Activities as Consistent Groups of Events

Dima Damen, David Hogg
2011 International Journal of Computer Vision  
Parsing a set of detections by such a grammar finds a consistent set of events that satisfies the activity's natural constraints. Each parse tree has a posterior probability in a Bayesian sense.  ...  The set of possible labellings of the Bayesian network corresponds to the set of all parse trees for a given set of detections.  ...  Values for the synthetic attribute t are assigned by the detector for each terminal symbol. Figure 3 shows two possible parse trees.  ... 
doi:10.1007/s11263-011-0497-0 fatcat:r7x7n44cyrbpzblfqeunzn5k4u

SSA-HIAST: A Novel Framework for Code Clone Detection

Neha Saini, Sukhdip Singh
2022 Computers Materials & Continua  
., Hashinspired abstract syntax trees.  ...  In the second phase, these abstract syntax trees are passed to a novel approach "Similarity-based self-adjusting hash inspired abstract syntax tree" algorithm that helps in knowing the similarity level  ...  The authors presented a parse tree kernel-based code plagiarism detection method. In terms of parse tree similarity, the parse tree kernel produces a similarity value between two source codes [20] .  ... 
doi:10.32604/cmc.2022.022659 fatcat:lh3i7dw62jeyxmobdd5vc2grpe

Strengthening Probabilistic Graphical Models: The Purge-and-merge Algorithm [article]

Simon Streicher, Johan du Preez
2021 arXiv   pre-print
The idea behind this algorithm is to iteratively nudge a malleable graph structure towards a tree structure by selectively merging factors.  ...  It is in principle possible to convert loopy PGMs to an equivalent tree structure, but for most interesting problems this is impractical due to exponential blow-up.  ...  By following our clustering procedure, we obtain the clusters C 1 , . . . , C m .  ... 
arXiv:2110.00091v1 fatcat:kgxcsqqwtjaota6fwhi7nfe45y

Statistical Parsing [chapter]

Stephen Clark
2010 The Handbook of Computational Linguistics and Natural Language Processing  
be used to compose trees for unseen sentences, motivated by the idea that important dependencies exist which cannot be captured by the usual head-based decomposition of a parse tree.  ...  Sampson's approach involved the use of manually parsed training material to obtain scores for local tree configurations, which were combined to obtain a score for a complete parse tree.  ... 
doi:10.1002/9781444324044.ch13 fatcat:psj22giwonaivd5zgaa2llj3ii
« Previous Showing results 1 — 15 out of 1,438 results