Image annotation using personal calendars as context Export

@inproceedings{citeulike:4098127, abstract = {In this paper, we introduce the idea of using the context of a personal calendar for labeling photo collections. Calendar event annotations are matched to images based on image capture time, and a Na\&\#239;ve Bayes model considers features from the calendar events as well as from computer vision-based image analysis to determine if the image actually matches the calendar event. This approach has the benefit that it requires no extra annotation from the consumer, since most people already keep calendars. In our test collections, 36\% of personal images could be tagged with a label from a personal calendar. Note that our preliminary results represent a lower bound on the performance that is possible because all the system components are expected to improve over time. As people migrate toward digital calendars, we can also expect more consistency in their calendar labels, which should improve the annotation accuracy.}, address = {New York, NY, USA}, author = {Gallagher, Andrew C. and Neustaedter, Carman G. and Cao, Liangliang and Luo, Jiebo and Chen, Tsuhan}, booktitle = {MM '08: Proceeding of the 16th ACM international conference on Multimedia}, citeulike-article-id = {4098127}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1459359.1459458}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1459359.1459458}, doi = {10.1145/1459359.1459458}, isbn = {978-1-60558-303-7}, keywords = {annotation, calendar, context, cvir}, location = {Vancouver, British Columbia, Canada}, pages = {681--684}, posted-at = {2009-02-25 12:30:17}, priority = {2}, publisher = {ACM}, title = {Image annotation using personal calendars as context}, url = {http://dx.doi.org/10.1145/1459359.1459458}, year = {2008} }

TY - CONF ID - citeulike:4098127 L3 - citeulike-article-id:4098127 N2 - In this paper, we introduce the idea of using the context of a personal calendar for labeling photo collections. Calendar event annotations are matched to images based on image capture time, and a Naïve Bayes model considers features from the calendar events as well as from computer vision-based image analysis to determine if the image actually matches the calendar event. This approach has the benefit that it requires no extra annotation from the consumer, since most people already keep calendars. In our test collections, 36% of personal images could be tagged with a label from a personal calendar. Note that our preliminary results represent a lower bound on the performance that is possible because all the system components are expected to improve over time. As people migrate toward digital calendars, we can also expect more consistency in their calendar labels, which should improve the annotation accuracy. CY - Vancouver, British Columbia, Canada SN - 978-1-60558-303-7 AD - New York, NY, USA EP - 684 TI - Image annotation using personal calendars as context PB - ACM T2 - MM '08: Proceeding of the 16th ACM international conference on Multimedia SP - 681 KW - annotation KW - calendar KW - context KW - cvir AU - Gallagher, Andrew AU - Neustaedter, Carman AU - Cao, Liangliang AU - Luo, Jiebo AU - Chen, Tsuhan PY - 2008/// UR - http://dx.doi.org/10.1145/1459359.1459458 ER -