Abnormal Event Detection in Videos using Generative Adversarial Nets. Ravanbakhsh, M., Nabi, M., Sangineto, E., Marcenaro, L., Regazzoni, C., & Sebe, N. arXiv:1708.09644 [cs], August, 2017. arXiv: 1708.09644![Abnormal Event Detection in Videos using Generative Adversarial Nets [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.
@article{ravanbakhsh_abnormal_2017,
title = {Abnormal {Event} {Detection} in {Videos} using {Generative} {Adversarial} {Nets}},
url = {http://arxiv.org/abs/1708.09644},
abstract = {In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.},
urldate = {2018-03-13TZ},
journal = {arXiv:1708.09644 [cs]},
author = {Ravanbakhsh, Mahdyar and Nabi, Moin and Sangineto, Enver and Marcenaro, Lucio and Regazzoni, Carlo and Sebe, Nicu},
month = aug,
year = {2017},
note = {arXiv: 1708.09644},
keywords = {Computer Science - Computer Vision and Pattern Recognition, Computer Science - Multimedia}
}
Downloads: 0
{"_id":"dyjrYs3EZuadWzMmm","bibbaseid":"ravanbakhsh-nabi-sangineto-marcenaro-regazzoni-sebe-abnormaleventdetectioninvideosusinggenerativeadversarialnets-2017","downloads":0,"creationDate":"2018-04-06T04:26:07.167Z","title":"Abnormal Event Detection in Videos using Generative Adversarial Nets","author_short":["Ravanbakhsh, M.","Nabi, M.","Sangineto, E.","Marcenaro, L.","Regazzoni, C.","Sebe, N."],"year":2017,"bibtype":"article","biburl":"https://bibbase.org/zotero/alwynmathew","bibdata":{"bibtype":"article","type":"article","title":"Abnormal Event Detection in Videos using Generative Adversarial Nets","url":"http://arxiv.org/abs/1708.09644","abstract":"In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.","urldate":"2018-03-13TZ","journal":"arXiv:1708.09644 [cs]","author":[{"propositions":[],"lastnames":["Ravanbakhsh"],"firstnames":["Mahdyar"],"suffixes":[]},{"propositions":[],"lastnames":["Nabi"],"firstnames":["Moin"],"suffixes":[]},{"propositions":[],"lastnames":["Sangineto"],"firstnames":["Enver"],"suffixes":[]},{"propositions":[],"lastnames":["Marcenaro"],"firstnames":["Lucio"],"suffixes":[]},{"propositions":[],"lastnames":["Regazzoni"],"firstnames":["Carlo"],"suffixes":[]},{"propositions":[],"lastnames":["Sebe"],"firstnames":["Nicu"],"suffixes":[]}],"month":"August","year":"2017","note":"arXiv: 1708.09644","keywords":"Computer Science - Computer Vision and Pattern Recognition, Computer Science - Multimedia","bibtex":"@article{ravanbakhsh_abnormal_2017,\n\ttitle = {Abnormal {Event} {Detection} in {Videos} using {Generative} {Adversarial} {Nets}},\n\turl = {http://arxiv.org/abs/1708.09644},\n\tabstract = {In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.},\n\turldate = {2018-03-13TZ},\n\tjournal = {arXiv:1708.09644 [cs]},\n\tauthor = {Ravanbakhsh, Mahdyar and Nabi, Moin and Sangineto, Enver and Marcenaro, Lucio and Regazzoni, Carlo and Sebe, Nicu},\n\tmonth = aug,\n\tyear = {2017},\n\tnote = {arXiv: 1708.09644},\n\tkeywords = {Computer Science - Computer Vision and Pattern Recognition, Computer Science - Multimedia}\n}\n\n","author_short":["Ravanbakhsh, M.","Nabi, M.","Sangineto, E.","Marcenaro, L.","Regazzoni, C.","Sebe, N."],"key":"ravanbakhsh_abnormal_2017","id":"ravanbakhsh_abnormal_2017","bibbaseid":"ravanbakhsh-nabi-sangineto-marcenaro-regazzoni-sebe-abnormaleventdetectioninvideosusinggenerativeadversarialnets-2017","role":"author","urls":{"Paper":"http://arxiv.org/abs/1708.09644"},"keyword":["Computer Science - Computer Vision and Pattern Recognition","Computer Science - Multimedia"],"downloads":0,"html":""},"search_terms":["abnormal","event","detection","videos","using","generative","adversarial","nets","ravanbakhsh","nabi","sangineto","marcenaro","regazzoni","sebe"],"keywords":["computer science - computer vision and pattern recognition","computer science - multimedia"],"authorIDs":[],"dataSources":["p3JdPh89hHfoARFkn"]}