Online multi-label active annotation: towards large-scale content-based video search Export

@inproceedings{Hua2008video-annotation, abstract = {Existing video search engines have not taken the advantages of video content analysis and semantic understanding. Video search in academia uses semantic annotation to approach content-based indexing. We argue this is a promising direction to enable real content-based video search. However, due to the complexity of both video data and semantic concepts, existing techniques on automatic video annotation are still not able to handle large-scale video set and large-scale concept set, in terms of both annotation accuracy and computation cost. To address this problem, in this paper, we propose a scalable framework for annotation-based video search, as well as a novel approach to enable large-scale semantic concept annotation, that is, online multi-label active learning. This framework is scalable to both the video sample dimension and concept label dimension. Large-scale unlabeled video samples are assumed to arrive consecutively in batches with an initial pre-labeled training set, based on which a preliminary multi-label classifier is built. For each arrived batch, a multi-label active learning engine is applied, which automatically selects and manually annotates a set of unlabeled sample-label pairs. And then an online learner updates the original classifier by taking the newly labeled sample-label pairs into consideration. This process repeats until all data are arrived. During the process, new labels, even without any pre-labeled training samples, can be incorporated into the process anytime. Experiments on TRECVID dataset demonstrate the effectiveness and efficiency of the proposed framework.}, address = {New York, NY, USA}, author = {Hua, Xian S. and Qi, Guo J.}, booktitle = {MM '08: Proceeding of the 16th ACM international conference on Multimedia}, citeulike-article-id = {3908878}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1459379}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1459359.1459379}, doi = {10.1145/1459359.1459379}, isbn = {978-1-60558-303-7}, keywords = {active, annotation, content, multi-label, online, search, video}, location = {Vancouver, British Columbia, Canada}, pages = {141--150}, posted-at = {2009-07-02 04:09:26}, priority = {2}, publisher = {ACM}, title = {Online multi-label active annotation: towards large-scale content-based video search}, url = {http://dx.doi.org/10.1145/1459359.1459379}, year = {2008} }

TY - CONF ID - Hua2008video-annotation L3 - citeulike-article-id:3908878 N2 - Existing video search engines have not taken the advantages of video content analysis and semantic understanding. Video search in academia uses semantic annotation to approach content-based indexing. We argue this is a promising direction to enable real content-based video search. However, due to the complexity of both video data and semantic concepts, existing techniques on automatic video annotation are still not able to handle large-scale video set and large-scale concept set, in terms of both annotation accuracy and computation cost. To address this problem, in this paper, we propose a scalable framework for annotation-based video search, as well as a novel approach to enable large-scale semantic concept annotation, that is, online multi-label active learning. This framework is scalable to both the video sample dimension and concept label dimension. Large-scale unlabeled video samples are assumed to arrive consecutively in batches with an initial pre-labeled training set, based on which a preliminary multi-label classifier is built. For each arrived batch, a multi-label active learning engine is applied, which automatically selects and manually annotates a set of unlabeled sample-label pairs. And then an online learner updates the original classifier by taking the newly labeled sample-label pairs into consideration. This process repeats until all data are arrived. During the process, new labels, even without any pre-labeled training samples, can be incorporated into the process anytime. Experiments on TRECVID dataset demonstrate the effectiveness and efficiency of the proposed framework. CY - Vancouver, British Columbia, Canada SN - 978-1-60558-303-7 AD - New York, NY, USA EP - 150 TI - Online multi-label active annotation: towards large-scale content-based video search PB - ACM T2 - MM '08: Proceeding of the 16th ACM international conference on Multimedia SP - 141 KW - active KW - annotation KW - content KW - multi-label KW - online KW - search KW - video AU - Hua, Xian AU - Qi, Guo PY - 2008/// UR - http://dx.doi.org/10.1145/1459359.1459379 ER -