action recognition using topic models

... IBM1 and IBM2 models, and inverse IBM1-inv and IBM2-inv models with the inverse dictionary from Eq However, a more interesting set up than using language models or translation models alone is ... Pattern Recognition, 43(5):1814– 1825 [Vidal et al.2006] E Vidal, F Casacuberta, L Rodr´guez, ı J Civera, and C Mart´nez 2006 Computer-assisted ı translation using speech recognition IEEE Transaction ... estimated using IBM models (Brown et al., 1993) The first term is an alignment probability while the second is a word dictionary Word dictionary probabilities can be directly estimated by IBM1 models...

Ngày tải lên: 07/03/2014, 22:20

... Classifier (K-NN) Action Action Action Action … … Detection Action (i) 2DPCA (Train or Test mode) fB Bin (B) layer Bin (2) layer Bin (1) layer Figure 1: The proposed human action recognition system ... W Hu, H Wang „Human action recognition using p ramid vocabular tree‟, ACCV, pp 527-537,Sep 2009 W Yang, Y Wang, and G Mori, “Human action recognition from a single clip per action , 2nd MLVMA ... and Pattern Recognition, Minneapolis, MN, USA, pp.1-8, Jun 2007 www.wisdom.weizmann.ac.il/~vision/SpaceTimeActions.html, Saad Ali, Mubarak Shah, "Human Action Recognition in Videos Using Kinematic...

Ngày tải lên: 24/04/2014, 12:57

... human actions Yilmaz and Shah [17] encode both shape and motion features to represent the 3D action models More specifically, they treat actions as 3D objects in (x, y, t) space and compute action ... Push-button ··· Query action (a) ··· Query action (b) Figure 12: Query action recognition for event retrieval (a) Turn-jump in ballet (b) Examples of recognizing pitching action First of all, ... of the query action recognition in the ballet sequence The turn-jump action is correctly detected among various jump actions as shown in Figure 12(a) In Figure 12(b), the pitching action is also...

Ngày tải lên: 21/06/2014, 18:20

... series For the action recognition problem, each BPAs series of testing action should be aligned with those of training action using DTW to result the value of similarity between to actions Table ... DTW algorithm for action similarity 3.2 Voting Algorithm for Action Recognition The distances between a testing action sample and all training action samples are obtained by using DTW It is clear ... parts stay still or take part in another action Two types of human actions are considered in this work in order to handle the task of action recognition Actions which are performed by motions of...

Ngày tải lên: 13/08/2015, 10:00

... translations using cross-lingual topic models have been made in the last few years, but they are model-dependent and not provide a general environment to adapt and apply other topic models for ... from latent topic models We have proven that topical knowledge is useful and improves the quality of word translations The quality of translations depends only on the quality of a topic model ... distribution for the BiLDA model, whereas this assumption clearly does not hold for topic models with a non-uniform topic prior P (w) is given by P (w) = P (w|Z)P (Z) If the assumption of uniformity...

Ngày tải lên: 23/03/2014, 16:20

... latent topics sentence -topic matrix auxiliary document -topic matrix 1: Randomly initialize U and V, and normalize them 2: repeat 3: Update U using Eq (3); 4: Update V using Eq (4); 5: Compute f using ... supV∈V f (U, V; Y) Instead of freely choosing topic unigram models, we further assume that topic unigram models are mixtures of some existing base unigram models, i.e = B(α) vkwd Ywd φvk;wd We have ... content of the document W is the W}i=1 vocabulary As topic models, we further assume the unigram model as a mixture of several topic unigram models, p(Wi |θ d ) = Definition The variational bound...

Ngày tải lên: 23/03/2014, 17:20

... Feature Maximum Extraction Selection Recognized Ơ Â Ơ U Ơ Ô ề ữ õ Â qĐàHàRTq%ảhề Block Viterbi Model Probability Computation Extraction Not Frontal Face Feature Extraction Probability ... W4ĐÂÊêuqH4CmĐpfêằHàHĐÊirĐ%Đơ ảÂ Â Ă Â ả Ơ Â Ư Â Â WuqĐãHCpf0huHẳ%qHHT ĐC Blocks Extraction Local Image Ă ÔĂ Ơ ỏ Ư ầ õ @ ụ õ ầ ò áơ ô ầ   ê i`qfƠ A4ê`rl%qfÂA!  f...

Ngày tải lên: 24/04/2014, 12:35

... Phoneme recognition using time–delay neural netwoks IEEE transactions on Acoustics, Speech and Signal Pro– cessing, 37:328–339, 1989 [12] T Watanabe and M Yachida Real–time gesture recognition using ... tem– poral pattern recognition Complex Systems, 3:349–381, 1989 [8] K Murakami and H Taguchi Gesture recognition using recurrent neural networks In Conference on Human Inter– action, page 237–242, ... 1986 [10] A Starner and T Pentland Visual recognition of American Sign Language using Hidden Markov Models In Int Conf on Automatic Face and Gesture Recognition, page 189–194, 1995 [11] A Waibel,...

Ngày tải lên: 24/04/2014, 12:54

... High Quality u Face Recognition in JPEG Compressed Images, In Proc IEEE Intern Conference on Image Processing, Kobe, 1999 [10] F.S Samaria, Face Recognition Using Hidden Markov Models, Ph D Thesis, ... results in a position and size invariant gesture recognition mode Summary Image sequence recognition based on novel pseudo three-dimensional Hidden Markov Models has been presented The modeling technique ... Ohya, and K Ishii, Recognizing Human Action in Time-Sequential Images Using Hidden Markov Model, In Proc IEEE Int Conference on Computer Vision and Pattern Recognition, 1992, pp 379385 [2] M Schuster...

Ngày tải lên: 24/04/2014, 13:44

... intraclass variabilities, Jhuang et al [22] perform actions recognition by building a neurobiological model using spatiotemporal gradient In [23], actions are recognized by training different SVM ... elaborated temporal dynamic models [24–27] Finally, there is also a fertile and broadly influential area of research that uses generative topic models for modeling and recognizing action categories based ... constraints of feature point are excluded 4.2 Local Feature Extraction Feature extraction forms the cornerstone of any action recognition procedure, but is also the most challenging and time-consuming...

Ngày tải lên: 21/06/2014, 07:20

... face recognition performed in both the spatial domain and using selected wavelet filters are presented in Table As can be seen from Table 1, the use of DWT for feature extraction improves recognition ... accuracy when used in a variety of face recognition approaches, and clearly this benefit extends to HMM-based face recognition Using Gabor filters increased recognition results even further The identification ... hidden markov models: application to handwritten numeral recognition, ” Intelligent Data Analysis Journal, vol 10, no 1, 2006 [23] D Bouchaffra and J Tan, “Structural hidden markov models using a relation...

Ngày tải lên: 22/06/2014, 06:20

Ngày tải lên: 06/09/2013, 11:10

... Workshop Yusuke Miyao 2005 Probabilistic disambiguation models for wide-coverage HPSG parsing In ACL ’05 Radford M Neal 1993 Probabilistic inference using markov chain Monte Carlo methods Technical ... Mihai Surdeanu, Sanda Harabagiu, John Williams, and Paul Aarseth 2003 Using predicate-argument structures for information extraction In ACL ’03 Adam R Teichert and Hal Daum´ III 2009 Unsupere vised ... on language-specific knowledge, either directly through heuristics, or indirectly through parsing models trained on treebanks Furthermore, some require labeled training instances for supervised...

Ngày tải lên: 19/02/2014, 19:20

... the probability under topic 1, topic 2, etc., or F2 : What is the probability under the most probable topic, second most, etc A model using F1 learns whether a specific topic is useful for translation, ... probabilistic membership This topic model infers the topic distribution of a test set and biases sentence translations to appropriate topics We accomplish this by introducing topic dependent lexical ... provides a topic distribution p(zn |di ) for zn = {1, , K} over each document, using Latent Dirichlet Allocation (LDA) (Blei et al., 2003) Then, we assign p(zn |di ) to be the topic distribution...

Ngày tải lên: 19/02/2014, 19:20

... UnigramDiac and UnigramLex are unigram models of the surface diacritized form and the lexeme respectively, and contain lexical information We also build 4gram lexeme models using an open-vocabulary language ... weighting of the features is one of the topics of this paper These 19 features consist of the 14 morphological features shown in Figure 1, which MADA predicts using 14 distinct Support Vector Machines ... (We also tested other ngram models, and found that a 4-gram lexeme model outperforms the other orders with n ≤ 5, although the improvement over the trigram and 5-gram models was less than 0.01%.)...

Ngày tải lên: 20/02/2014, 09:20

... Dialogue Speech Recognition using Cross-word Context Constrained Word Graphs ICASSP 96, pp 145-148, 1996 Y Wakita et al., 97 Correct parts extraction from speech recognition results using semantic ... Results of Speech Recognition: We used 4806 recognition results including errors, from the output of speech recognition (Masataki et al., 96; Shimizu et al., 96) experiment using an ATR spoken ... not less than 2, therefore we obtained 629 Error-Pattems using the 4321 results of speech recognition Preparation of the String-Database: Using the different data-sets of the ATR spoken language...

Ngày tải lên: 20/02/2014, 18:20

... in applying probabilistic models to natural language is deciding what aspects of the sentence and the discourse are relevant to the model Most previous probabilistic models of parsing assume the ... of these models come close to incorporating enough context to disambiguate many cases of ambiguity A significant reason researchers have limited the contextual information used by their models ... for this model using decision trees The history-based grammar model provides a mechanism for taking advantage of contextual information from anywhere in the discourse history Using decision tree...

Ngày tải lên: 20/02/2014, 21:20

... adopting topic models here To employ topic models, we treat RASCs as “documents”, items as “words”, and the final semantic classes as “topics” There are, however, several challenges in applying topic ... scope of topic modeling Topic Models In this section we briefly introduce the two widely used topic models which are adopted in our paper Both of them model a document as a mixture of hidden topics ... for details) before building topic models for CR(q), where some lowfrequency items are removed Determine the number of topics: Most topic models require the number of topics to be known beforehand1...

Ngày tải lên: 08/03/2014, 00:20

... Nikravesh, and L Zadeh, editors 2006 Feature Extraction, Foundations and Applications Springer References S Harabagiu and A Hickl 2006 Methods for using textual entailment in open-domain question ... information extraction by multi-level bootstrapping In Proceedings of AAAI, pages 474–479, Orlando, FL Y Shinyama and S Sekine 2003 Paraphrase acquisition for information extraction In Proceedings ... returned by similarity measures that indicate how similar Si,1 and Si,2 are at various levels of abstraction The vectors and the corresponding categories { v1 , yi , , , yn } are given as input to...

Ngày tải lên: 08/03/2014, 01:20