{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,21]],"date-time":"2026-01-21T22:09:12Z","timestamp":1769033352863,"version":"3.49.0"},"reference-count":38,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,22]]},"DOI":"10.1109\/cbmi66578.2025.11339305","type":"proceedings-article","created":{"date-parts":[[2026,1,20]],"date-time":"2026-01-20T20:38:56Z","timestamp":1768941536000},"page":"1-8","source":"Crossref","is-referenced-by-count":0,"title":["Zero-Shot Vision-Language Model for Event Detection in Smart Surveillance"],"prefix":"10.1109","author":[{"given":"Younes","family":"Kebour","sequence":"first","affiliation":[{"name":"LAMIH CNRS, Univ. Polytechnique Hauts-de-France Junia, Grande École d'Ingénieur,Lille,France"}]},{"given":"Smail","family":"Niar","sequence":"additional","affiliation":[{"name":"Univ. Polytechnique Hauts-de-France,LAMIH CNRS INSA Hauts-de-France,Valenciennes,France"}]},{"given":"Nacim","family":"Ihaddadene","sequence":"additional","affiliation":[{"name":"Junia, Grande École d'Ingénieur,Lille,France"}]},{"given":"Abdelghani","family":"Bekrar","sequence":"additional","affiliation":[{"name":"LAMIH CNRS, Univ. Polytechnique Hauts-de-France,Valenciennes,France"}]},{"given":"Hammouda","family":"Elbez","sequence":"additional","affiliation":[{"name":"CITC-EuraRFID,Lille,France"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/wacv57701.2024.00020"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-022-01578-9"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01953"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02080"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3074805"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW63382.2024.00202"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i3.27963"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/DSC54232.2022.9888874"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01878"},{"key":"ref10","article-title":"Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision","volume-title":"Proceedings of the 38th International Conference on Ma-chine Learning","author":"Jia"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/6979.880968"},{"key":"ref12","article-title":"BRAVE: Broadening the vi-sual encoding of vision-language models","author":"Kar","journal-title":"arXiv preprint"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-023-10595-0"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.5220\/0011663600003417"},{"key":"ref15","article-title":"BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation","volume-title":"International Conference on Machine Learning","author":"Li"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/WACV57701.2024.00113"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3609333"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02347"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3072863"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3072863"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00775"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2010.5539872"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46478-7_1"},{"key":"ref24","volume-title":"ImaginaryNet: Learning Object Detectors without Real Images and Annotations","author":"Ni"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2024.3451935"},{"key":"ref26","volume-title":"Learning Transferable Visual Models From Natural Language Supervision","author":"Radford"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/IC3S57698.2023.10169381"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00678"},{"key":"ref29","volume-title":"Attention is All you Need","author":"Vaswani"},{"key":"ref30","article-title":"GIT: A Generative Image-to-text Transformer for Vision and Language","author":"Wang","journal-title":"Transactions on Machine Learning Research"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3062192"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58577-8_20"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01732"},{"key":"ref34","article-title":"mPLUG-2: a modularized multi-modal foundation model across text, image and video","volume-title":"Proceedings of the 40th International Conference on Machine Learning. JMLR.org","author":"Xu"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02082"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3369699"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3450734"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/s13735-023-00267-8"}],"event":{"name":"2025 International Conference on Content-Based Multimedia Indexing (CBMI)","location":"Dublin, Ireland","start":{"date-parts":[[2025,10,22]]},"end":{"date-parts":[[2025,10,24]]}},"container-title":["2025 International Conference on Content-Based Multimedia Indexing (CBMI)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11339229\/11339242\/11339305.pdf?arnumber=11339305","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,21]],"date-time":"2026-01-21T07:24:58Z","timestamp":1768980298000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11339305\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":38,"URL":"https:\/\/doi.org\/10.1109\/cbmi66578.2025.11339305","relation":{},"subject":[],"published":{"date-parts":[[2025,10,22]]}}}