{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T18:10:24Z","timestamp":1790187024080,"version":"4.1.0"},"reference-count":44,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>With smart gadgets and computers everywhere, real-time video monitoring is now necessary to keep people secure and lower the chance of unexpected human actions. However, current surveillance systems have a lot of trouble effectively spotting unusual actions in real life. Most of the advanced anomaly detection algorithms are trained and tested on synthetic or simulated video datasets. These datasets frequently do not show how complicated and different real-life settings can be. Because of this, their usefulness and capacity to be used in real life are still restricted. We propose a transfer learning-based hybrid deep framework for finding unusual activity in real-world surveillance. We test the model using raw video streams from a subset of the UCF-Crime dataset that shows actual situations of people doing things. First, a pre-trained DenseNet-201, a convolutional neural network, extracts spatial data from video frames that have already been processed. A Bidirectional Long Short-Term Memory (Bi-LSTM) network is used to describe temporal relationships between video sequences in a useful way. The spatial data and the Bi-LSTM\u2019s temporal representations are then integrated to make anomaly identification more precise and dependable. This integrated spatiotemporal architecture identifies abnormal actions of human beings in challenging monitoring situations. The model that was recommended performs better than the state of art methods. It gets 95.04% accuracy on the UCF50 dataset and 62.04% on the untrimmed UCF-Crime dataset. These results suggest that the model performs effective in real-life scenarios.<\/jats:p>","DOI":"10.1371\/journal.pone.0355690","type":"journal-article","created":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T17:53:27Z","timestamp":1790186007000},"page":"e0355690","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["A spatiotemporal deep learning framework for detecting unusual human activity in surveillance videos"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5259-4208","authenticated-orcid":true,"given":"Manoj","family":"Kumar","sequence":"first","affiliation":[{"name":"Bennett University","place":["Greater Noida, India"],"department":["School of Computer Science Engineering & Technology"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9027-4432","authenticated-orcid":true,"given":"Mantosh","family":"Biswas","sequence":"additional","affiliation":[{"name":"North Campus, University of Delhi","place":["Delhi, India"],"department":["Department of Computer Science"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anoop Kumar","family":"Patel","sequence":"additional","affiliation":[{"name":"National Institute of Technology","place":["Kurukshetra, Haryana, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Abhay","family":"Kumar","sequence":"additional","affiliation":[{"name":"NIT Patna","place":["Patna, Bihar, India"],"department":["Department of Computer Science and Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kumar","family":"Abhishek","sequence":"additional","affiliation":[{"name":"NIT Patna","place":["Patna, Bihar, India"],"department":["Department of Computer Science and Engineering"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6555-6980","authenticated-orcid":true,"given":"Ahamed","family":"Shafeeq B. M.","sequence":"additional","affiliation":[{"name":"Manipal Institute of Technology, Manipal Academy of Higher Education","place":["Manipal, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,23]]},"reference":[{"issue":"4","key":"pone.0355690.ref001","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0321754","article-title":"Intelligent recognition of human activities using deep learning techniques","volume":"20","author":"S Bashir","year":"2025","journal-title":"PLoS One"},{"key":"pone.0355690.ref002","doi-asserted-by":"crossref","first-page":"103126","DOI":"10.1016\/j.cviu.2020.103126","article-title":"Embedding group and obstacle information in LSTM networks for human trajectory prediction in crowded scenes","volume":"203","author":"N Bisagno","year":"2021","journal-title":"Computer Vision and Image Understanding"},{"issue":"3","key":"pone.0355690.ref003","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0118723","article-title":"Step detection and activity recognition accuracy of seven physical activity monitors","volume":"10","author":"FA Storm","year":"2015","journal-title":"PLoS One"},{"issue":"2","key":"pone.0355690.ref004","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0228869","article-title":"Convolutional and recurrent neural network for human activity recognition: Application on American sign language","volume":"15","author":"V Hernandez","year":"2020","journal-title":"PLoS One"},{"issue":"1","key":"pone.0355690.ref005","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0262181","article-title":"A deep neural network model for multi-view human activity recognition","volume":"17","author":"PU Putra","year":"2022","journal-title":"PLoS One"},{"key":"pone.0355690.ref006","doi-asserted-by":"crossref","unstructured":"Baisware A, Sayankar B, Hood S. Review on Recent Advances in Human Action Recognition in Video Data. In: 2019 9th International Conference on Emerging Trends in Engineering and Technology - Signal and Information Processing (ICETET-SIP-19), 2019. 1\u20135. https:\/\/doi.org\/10.1109\/icetet-sip-1946815.2019.9092193","DOI":"10.1109\/ICETET-SIP-1946815.2019.9092193"},{"issue":"7","key":"pone.0355690.ref007","doi-asserted-by":"crossref","first-page":"4507","DOI":"10.1109\/TITS.2020.3017882","article-title":"Novel deep learning-enabled LSTM autoencoder architecture for discovering anomalous events from intelligent transportation systems","volume":"22","author":"J Ashraf","year":"2021","journal-title":"IEEE Trans Intell Transport Syst"},{"key":"pone.0355690.ref008","doi-asserted-by":"crossref","first-page":"103232","DOI":"10.1016\/j.jvcir.2021.103232","article-title":"A comparative study between single and multi-frame anomaly detection and localization in recorded video streams","volume":"79","author":"M Bahrami","year":"2021","journal-title":"Journal of Visual Communication and Image Representation"},{"key":"pone.0355690.ref009","doi-asserted-by":"crossref","unstructured":"Sultani W, Chen C, Shah M. Real-World Anomaly Detection in Surveillance Videos. In: 2018 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, 2018. 6479\u201388. https:\/\/doi.org\/10.1109\/cvpr.2018.00678","DOI":"10.1109\/CVPR.2018.00678"},{"issue":"3","key":"pone.0355690.ref010","doi-asserted-by":"crossref","first-page":"313","DOI":"10.1109\/TSMCC.2011.2149519","article-title":"Hierarchical filtered motion for action recognition in crowded videos","volume":"42","author":"Y Tian","year":"2012","journal-title":"IEEE Trans Syst, Man, Cybern C"},{"key":"pone.0355690.ref011","author":"C Harris","year":"2013"},{"issue":"8","key":"pone.0355690.ref012","doi-asserted-by":"crossref","first-page":"1422","DOI":"10.1109\/TCSVT.2013.2244794","article-title":"Action recognition using multilevel features and latent structural SVM","volume":"23","author":"X Wu","year":"2013","journal-title":"IEEE Trans Circuits Syst Video Technol"},{"issue":"12","key":"pone.0355690.ref013","doi-asserted-by":"crossref","first-page":"2247","DOI":"10.1109\/TPAMI.2007.70711","article-title":"Actions as space-time shapes","volume":"29","author":"L Gorelick","year":"2007","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0355690.ref014","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":"D Weinland","year":"2006","journal-title":"Computer Vision and Image Understanding"},{"key":"pone.0355690.ref015","doi-asserted-by":"crossref","unstructured":"Laptev I, Lindeberg T. Space-time interest points. In: Proceedings Ninth IEEE International Conference on Computer Vision, 2003. 432\u20139 vol.1. https:\/\/doi.org\/10.1109\/iccv.2003.1238378","DOI":"10.1109\/ICCV.2003.1238378"},{"key":"pone.0355690.ref016","doi-asserted-by":"crossref","unstructured":"Willems G, Tuytelaars T, An LG (2008) Efficient Dense and Scale-Invariant Spatio-Temporal. 650\u201363.","DOI":"10.1007\/978-3-540-88688-4_48"},{"key":"pone.0355690.ref017","doi-asserted-by":"crossref","unstructured":"Lin T-Y, Dollar P, Girshick R, He K, Hariharan B, Belongie S. Feature Pyramid Networks for Object Detection. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. 936\u201344. https:\/\/doi.org\/10.1109\/cvpr.2017.106","DOI":"10.1109\/CVPR.2017.106"},{"key":"pone.0355690.ref018","doi-asserted-by":"crossref","unstructured":"Wang H, Ullah MM, Klaser A, Laptev I, Schmid C. Evaluation of local spatio-temporal features for action recognition. In: Procedings of the British Machine Vision Conference 2009, 2009. 124.1-124.11. https:\/\/doi.org\/10.5244\/c.23.124","DOI":"10.5244\/C.23.124"},{"key":"pone.0355690.ref019","doi-asserted-by":"crossref","unstructured":"Laptev I, Marszalek M, Schmid C, Rozenfeld B. Learning realistic human actions from movies. In: 2008 IEEE Conference on Computer Vision and Pattern Recognition, 2008. 1\u20138. https:\/\/doi.org\/10.1109\/cvpr.2008.4587756","DOI":"10.1109\/CVPR.2008.4587756"},{"key":"pone.0355690.ref020","first-page":"428","volume-title":"Human Detection Using Oriented Histograms of Flow and Appearance To cite this version: Human Detection using Oriented Histograms of Flow and Appearance. Eur Conf Comput Vis (ECCV \u201906)","author":"N Dalal","year":"2006"},{"key":"pone.0355690.ref021","first-page":"211201686","article-title":"Make A Long Image Short: Adaptive Token Length for Vision Transformers","author":"Y Zhu","year":"2021","journal-title":"arXiv Prepr arXiv."},{"key":"pone.0355690.ref022","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":"AF Bobick","year":"2001","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0355690.ref023","doi-asserted-by":"crossref","unstructured":"Umakanthan S, Denman S, Sridharan S, Fookes C, Wark T. Spatio Temporal Feature Evaluation for Action Recognition. In: 2012 International Conference on Digital Image Computing Techniques and Applications (DICTA), 2012. 1\u20138. https:\/\/doi.org\/10.1109\/dicta.2012.6411720","DOI":"10.1109\/DICTA.2012.6411720"},{"issue":"8","key":"pone.0355690.ref024","doi-asserted-by":"crossref","first-page":"1576","DOI":"10.1109\/TPAMI.2011.253","article-title":"Learning sparse representations for human action recognition","volume":"34","author":"T Guha","year":"2012","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"pone.0355690.ref025","article-title":"DeepUSPS: Deep robust unsupervised saliency prediction with self-supervision","author":"DT Nguyen","year":"2019","journal-title":"Adv Neural Inf Process Syst 32"},{"key":"pone.0355690.ref026","doi-asserted-by":"crossref","unstructured":"Jia C, Kong Y, Ding Z, Fu YR. Latent Tensor Transfer Learning for RGB-D Action Recognition. In: Proceedings of the 22nd ACM international conference on Multimedia, 2014. 87\u201396. https:\/\/doi.org\/10.1145\/2647868.2654928","DOI":"10.1145\/2647868.2654928"},{"key":"pone.0355690.ref027","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1016\/j.neucom.2016.03.024","article-title":"Activity recognition using a supervised non-parametric hierarchical HMM","volume":"199","author":"N Raman","year":"2016","journal-title":"Neurocomputing"},{"key":"pone.0355690.ref028","doi-asserted-by":"crossref","unstructured":"Shiranthika C, Premakumara N, Chiu H-L, Samani H, Shyalika C, Yang C-Y. Human Activity Recognition Using CNN & LSTM. In: 2020 5th International Conference on Information Technology Research (ICITR), 2020. 1\u20136. https:\/\/doi.org\/10.1109\/icitr51448.2020.9310792","DOI":"10.1109\/ICITR51448.2020.9310792"},{"key":"pone.0355690.ref029","doi-asserted-by":"crossref","unstructured":"Hasan M, Choi J, Neumann J, Roy-Chowdhury AK, Davis LS. Learning Temporal Regularity in Video Sequences. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. 733\u201342. https:\/\/doi.org\/10.1109\/cvpr.2016.86","DOI":"10.1109\/CVPR.2016.86"},{"key":"pone.0355690.ref030","doi-asserted-by":"crossref","first-page":"687","DOI":"10.1016\/j.future.2021.07.015","article-title":"Deep transfer learning framework for the identification of malicious activities to combat cyberattack","volume":"125","author":"D Singh","year":"2021","journal-title":"Futur Gener Comput Syst"},{"key":"pone.0355690.ref031","doi-asserted-by":"crossref","unstructured":"Jodoin P-M, Konrad J, Saligrama V. Modeling background activity for behavior subtraction. In: 2008 Second ACM\/IEEE International Conference on Distributed Smart Cameras, 2008. 1\u201310. https:\/\/doi.org\/10.1109\/icdsc.2008.4635683","DOI":"10.1109\/ICDSC.2008.4635683"},{"key":"pone.0355690.ref032","first-page":"1","article-title":"International Transaction Journal of Engineering, Management","volume":"11","author":"AJ Alzahrani","year":"2020","journal-title":"& Applied Sciences & Technologies Anomaly Detection in Crowds by Fusion of Novel 1 Introduction"},{"key":"pone.0355690.ref033","doi-asserted-by":"crossref","first-page":"543","DOI":"10.1016\/j.neucom.2015.11.049","article-title":"Analyzing crowd behavior in naturalistic conditions: Identifying sources and sinks and characterizing main flows","volume":"177","author":"SD Khan","year":"2016","journal-title":"Neurocomputing"},{"issue":"7","key":"pone.0355690.ref034","doi-asserted-by":"crossref","first-page":"12565","DOI":"10.1109\/TNNLS.2024.3449573","article-title":"Anomaly detection in hyperspectral images using adaptive graph frequency location","volume":"36","author":"B Tu","year":"2025","journal-title":"IEEE Trans Neural Netw Learn Syst"},{"issue":"2","key":"pone.0355690.ref035","doi-asserted-by":"crossref","first-page":"2190","DOI":"10.1109\/TII.2023.3280337","article-title":"A multihead attention self-supervised representation model for industrial sensors anomaly detection","volume":"20","author":"Y Qiao","year":"2024","journal-title":"IEEE Trans Ind Inf"},{"key":"pone.0355690.ref036","doi-asserted-by":"crossref","first-page":"5115","DOI":"10.1109\/TIP.2025.3595408","article-title":"Multi-scale autoencoder suppression strategy for hyperspectral image anomaly detection","volume":"34","author":"B Tu","year":"2025","journal-title":"IEEE Trans Image Process"},{"key":"pone.0355690.ref037","first-page":"2987","article-title":"Robust Human Interaction Recognition Using Extended Kalman Filter","volume":"81","author":"TFN Bukht","year":"2024","journal-title":"Comput Mater Contin"},{"issue":"18","key":"pone.0355690.ref038","doi-asserted-by":"crossref","first-page":"55981","DOI":"10.1007\/s11042-023-17748-x","article-title":"Real-time detection of abnormal human activity using deep learning and temporal attention mechanism in video surveillance","volume":"83","author":"M Kumar","year":"2023","journal-title":"Multimed Tools Appl"},{"issue":"22","key":"pone.0355690.ref039","doi-asserted-by":"crossref","first-page":"61843","DOI":"10.1007\/s11042-023-15904-x","article-title":"Abnormal human activity detection by convolutional recurrent neural network using fuzzy logic","volume":"83","author":"M Kumar","year":"2023","journal-title":"Multimed Tools Appl"},{"key":"pone.0355690.ref040","doi-asserted-by":"crossref","unstructured":"Ullah A, Muhammad K, Haydarov K, Haq IU, Lee M, Baik SW. One-Shot Learning for Surveillance Anomaly Recognition using Siamese 3D CNN. In: 2020 International Joint Conference on Neural Networks (IJCNN), 2020. 1\u20138. https:\/\/doi.org\/10.1109\/ijcnn48605.2020.9207595","DOI":"10.1109\/IJCNN48605.2020.9207595"},{"key":"pone.0355690.ref041","doi-asserted-by":"crossref","unstructured":"Sun L, Chen Y, Luo W, Wu H, Zhang C. Discriminative Clip Mining for Video Anomaly Detection. In: 2020 IEEE International Conference on Image Processing (ICIP), 2020. 2121\u20135. https:\/\/doi.org\/10.1109\/icip40778.2020.9191072","DOI":"10.1109\/ICIP40778.2020.9191072"},{"key":"pone.0355690.ref042","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1016\/j.patrec.2017.12.024","article-title":"A Bag of Expression framework for improved human action recognition","volume":"103","author":"S Nazir","year":"2018","journal-title":"Pattern Recognition Letters"},{"issue":"12","key":"pone.0355690.ref043","doi-asserted-by":"crossref","first-page":"9634","DOI":"10.1109\/JIOT.2020.3027483","article-title":"Multiview summarization and activity recognition meet edge computing in IoT environments","volume":"8","author":"T Hussain","year":"2021","journal-title":"IEEE Internet Things J"},{"key":"pone.0355690.ref044","first-page":"1","article-title":"LRCN Based human activity recognition from video data","volume":"28","author":"MS Uzzaman","year":"2022","journal-title":"ssrn"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0355690","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T17:53:36Z","timestamp":1790186016000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0355690"}},"subtitle":[],"editor":[{"given":"Jin","family":"Liu","sequence":"first","affiliation":[{"name":"Shanghai Maritime University","place":["CHINA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,23]]},"references-count":44,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,23]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0355690","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,23]]}}}