{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T13:47:50Z","timestamp":1760708870475,"version":"build-2065373602"},"reference-count":41,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2015,7,16]],"date-time":"2015-07-16T00:00:00Z","timestamp":1437004800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Ambient Assisted Living Joint Programme","award":["AAL-JP-2012-5-045"],"award-info":[{"award-number":["AAL-JP-2012-5-045"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Multi-view action recognition has gained a great interest in video surveillance, human computer interaction, and multimedia retrieval, where multiple cameras of different types are deployed to provide a complementary field of views. Fusion of multiple camera views evidently leads to more robust decisions on both tracking multiple targets and analysing complex human activities, especially where there are occlusions. In this paper, we incorporate the marginalised stacked denoising autoencoders (mSDA) algorithm to further improve the bag of words (BoWs) representation in terms of robustness and usefulness for multi-view action recognition. The resulting representations are fed into three simple fusion strategies as well as a multiple kernel learning algorithm at the classification stage. Based on the internal evaluation, the codebook size of BoWs and the number of layers of mSDA may not significantly affect recognition performance. According to results on three multi-view benchmark datasets, the proposed framework improves recognition performance across all three datasets and outputs record recognition performance, beating the state-of-art algorithms in the literature. It is also capable of performing real-time action recognition at a frame rate ranging from 33 to 45, which could be further improved by using more powerful machines in future applications.<\/jats:p>","DOI":"10.3390\/s150717209","type":"journal-article","created":{"date-parts":[[2015,7,16]],"date-time":"2015-07-16T10:11:44Z","timestamp":1437041504000},"page":"17209-17231","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":10,"title":["Marginalised Stacked Denoising Autoencoders for Robust Representation of Real-Time Multi-View Action Recognition"],"prefix":"10.3390","volume":"15","author":[{"given":"Feng","family":"Gu","sequence":"first","affiliation":[{"name":"School of Computing and Information Systems, Kingston University, Penrhyn Road, Kingston upon Thames KT1 2EE, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3391-711X","authenticated-orcid":false,"given":"Francisco","family":"Fl\u00f3rez-Revuelta","sequence":"additional","affiliation":[{"name":"Faculty of Science, Engineering and Computing, Kingston University, Penrhyn Road, Kingston upon Thames KT1 2EE, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dorothy","family":"Monekosso","sequence":"additional","affiliation":[{"name":"School of Computing and Information Systems, Kingston University, Penrhyn Road, Kingston upon Thames KT1 2EE, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paolo","family":"Remagnino","sequence":"additional","affiliation":[{"name":"School of Computing and Information Systems, Kingston University, Penrhyn Road, Kingston upon Thames KT1 2EE, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2015,7,16]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Laptev, I., Marszalek, M., Schmid, C., and Rozenfeld, B. (2008, January 23\u201328). Learning realistic human actions from movies. Anchorage, AK, USA.","DOI":"10.1109\/CVPR.2008.4587756"},{"key":"ref_2","unstructured":"MacKay, D. (2003). Information Theory, Inference, and Learning Algorithms, Cambridge University Press."},{"key":"ref_3","unstructured":"Xu, Z., Chen, M., Weinberger, K., and Sha, F. (November, January 29). From sBoW to dCoT: Marginalized Encoders for Text Representationn. Maui, HI, USA."},{"key":"ref_4","unstructured":"Linsker, R. (2008). Advances in Neural Information Processing Systems (NIPS), Morgan Kaufmann Publishers Inc."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Gu, F., Florez-Revuelta, F., Monekosso, D., and Remagnino, P. (2014, January 2\u20135). A Multiple Kernel Learning Based Fusion Framework for Real-Time Multi-View Action Recognition. Belfast, UK.","DOI":"10.1007\/978-3-319-13105-4_5"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1016\/j.cviu.2006.07.013","article-title":"Free viewpoint action recognition using motion history volumes","volume":"104","author":"Weinland","year":"2006","journal-title":"Comput. Vis. Image Underst."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1109\/34.910878","article-title":"The recognition of human movement using temporal templates","volume":"23","author":"Bobick","year":"2001","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell. PAMI."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"78","DOI":"10.1016\/j.neucom.2011.03.051","article-title":"A probabilistic, discriminative and distributed system for the recognition of human actions from multiple views","volume":"75","author":"Cilla","year":"2012","journal-title":"Neurocomputing"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"354","DOI":"10.1111\/exsy.12040","article-title":"Human action recognition with sparse classification and multiple-view learning","volume":"31","author":"Cilla","year":"2013","journal-title":"Expert Syst."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"8750","DOI":"10.3390\/s130708750","article-title":"Multi-View Human Activity Recognition in Distributed Camera Sensor Networks","volume":"13","author":"Adeli","year":"2013","journal-title":"Sensors"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Burghouts, G., Eendebak, P., Bouma, H., and ten Hove, J.M. (2013, January 27\u201330). Improved action recognition by combining multiple 2D views in the bag-of-words model. Krakow, Poland.","DOI":"10.1109\/AVSS.2013.6636648"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"553","DOI":"10.1109\/JSTSP.2012.2193556","article-title":"A local 3-D motion descriptor for mult-view human action recognition from 4-D spatio-temporal interest points","volume":"6","author":"Holte","year":"2012","journal-title":"IEEE J. Sel. Top. Signal Process."},{"key":"ref_13","unstructured":"Yan, P., Khan, S., and Shah, M. (2008, January 23\u201328). Learning 4D action feature models for arbitrary view action recognition. Anchorage, AK, USA."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"347","DOI":"10.1016\/j.cviu.2011.08.008","article-title":"Multi-view human movement recognition based on fuzzy distances and linear discriminant analysis","volume":"116","author":"Iosifidis","year":"2012","journal-title":"Comput. Vis. Image Underst."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Hashemi, S., and Rahmati, M. (2015). View-independent action recognition: A hybrid approach. Multimed. Tools Appl.","DOI":"10.1007\/s11042-015-2606-5"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1016\/j.patrec.2012.04.016","article-title":"Multi-view action recognition using local similarity random forests and sensor fusion (Extracting Semantics from Multi-Spectrum Video)","volume":"34","author":"Zhu","year":"2013","journal-title":"Pattern Recognit. Lett."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Iosifidis, A., Tefas, A., and Pitas, I. (2012, January 10\u201315). Neural representation and learning for multi-view human action recognition. Brisbane, Australia.","DOI":"10.1109\/IJCNN.2012.6252675"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Weinland, D., \u00d6zuysal, M., and Fua, P. (2010, January 5\u201311). Making Action Recognition Robust to Occlusions and Viewpoint Changes. Crete, Greece.","DOI":"10.1007\/978-3-642-15558-1_46"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Cherla, S., Kulkarni, K., Kale, A., and Ramasubramanian, V. (2008, January 23\u201328). Towards fast, view-invariant human action recognition. Anchorage, AK, USA.","DOI":"10.1109\/CVPRW.2008.4563179"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1799","DOI":"10.1016\/j.patrec.2013.01.021","article-title":"Silhouette-based human action recognition using sequences of key poses","volume":"34","author":"Chaaraoui","year":"2013","journal-title":"Pattern Recognit. Lett."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"8895","DOI":"10.3390\/s140508895","article-title":"A vision-based system for intelligent monitoring: Human behaviour analysis and Privacy by context","volume":"14","author":"Chaaraoui","year":"2014","journal-title":"Sensors"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Wang, H., and Schmid, C. (2013, January 1\u20138). Action Recognition with Improved Trajectories. Sydney, Australia.","DOI":"10.1109\/ICCV.2013.441"},{"key":"ref_23","unstructured":"Bishop, C. (2006). Pattern Recognition and Machine Learning, Springer."},{"key":"ref_24","first-page":"3371","article-title":"Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion","volume":"11","author":"Vincent","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref_25","unstructured":"Bengio, Y., Lamblin, P., Popovici, D., and Larochelle, H. (2007). Advances in Neural Information Processing Systems (NIPS), MIT Press."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"key":"ref_27","first-page":"1","article-title":"Exploring strategies for training deep neural networks","volume":"10","author":"Larochelle","year":"2009","journal-title":"J. Mach. Learn. Res."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Blitzer, J., McDonald, R., and Pereira, F. (2006, January 22\u201323). Domain adaptation with structural correspondence learning. Sydney, Australia.","DOI":"10.3115\/1610075.1610094"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Vincent, P., Larochelle, H., Bengio, Y., and Manzagol, P.A. (2008, January 5\u20139). Extracting and composing robust features with denoising autoencoders. Helsinki, Finland.","DOI":"10.1145\/1390156.1390294"},{"key":"ref_30","unstructured":"Bergstra, J., Breuleux, O., Bastien, F., Lamblin, P., Pascanu, R., Desjardins, G., Turian, J., Warde-Farley, D., and Bengio, Y. (July, January 28). Theano: A CPU and GPU Math Expression Compiler. Austin, TX, USA."},{"key":"ref_31","unstructured":"Dauphin, Y., Glorot, X., and Bengio, Y. (July, January 28). Large-Scale Learning of Embeddings with Reconstruction Sampling. Bellevue, WA, USA."},{"key":"ref_32","unstructured":"Chen, M., Xu, Z., Weinberger, K., and Sha, F. (2012, January 3\u20136). Marginalized Stacked Denoising Autoencoders. Utah, UT, USA."},{"key":"ref_33","unstructured":"Chen, M., Xu, Z., Weinberger, K., and Sha, F. (July, January 26). Marginalized Stacked Denoising Autoencoders for Domain Adaptation. Edinburgh, Scotland, UK."},{"key":"ref_34","unstructured":"Glorot, X., Bordes, A., and Bengio, Y. (July, January 28). Domain adaptation for large-scale sentiment classification: A deep learning approach. Bellevue, Washington, WA, USA."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"226","DOI":"10.1109\/34.667881","article-title":"On combining classifiers","volume":"20","author":"Kittler","year":"1998","journal-title":"IEEE Trans Pattern Anal. Mach. Intell."},{"key":"ref_36","first-page":"2211","article-title":"Multiple Kernel Learning Algorithms","volume":"12","author":"Gonen","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_37","unstructured":"Vedaldi, A., Gulshan, V., Varma, M., and Zisserman, A. (October, January 27). Multiple Kernels for Object Detection. Kyoto, Japan."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Varma, M., and Babu, B.R. (2009, January 14\u201318). More generality in efficient multiple kernel learning. Montreal, QC, Canada.","DOI":"10.1145\/1553374.1553510"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Cheng, Z., Qin, L., Ye, Y., Huang, Q., and Tian, Q. (2012, January 7\u201313). Human Daily Action Analysis with Multi-view and Color-Depth Data. Firenze, Italy.","DOI":"10.1007\/978-3-642-33868-7_6"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/1961189.1961199","article-title":"LIBSVM: A library for support vector machines","volume":"2","author":"Chang","year":"2011","journal-title":"ACM Trans. Intell. Syst. Technol."},{"key":"ref_41","doi-asserted-by":"crossref","unstructured":"Ni, B., Wang, G., and Moulin, P. (2011, January 12). RGBD-HUDAAC: A color-depth video database for human daily activity recognition. Barcelona, Spain.","DOI":"10.1109\/ICCVW.2011.6130379"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/15\/7\/17209\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T20:49:15Z","timestamp":1760215755000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/15\/7\/17209"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2015,7,16]]},"references-count":41,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2015,7]]}},"alternative-id":["s150717209"],"URL":"https:\/\/doi.org\/10.3390\/s150717209","relation":{},"ISSN":["1424-8220"],"issn-type":[{"type":"electronic","value":"1424-8220"}],"subject":[],"published":{"date-parts":[[2015,7,16]]}}}