TY - GEN
T1 - Building text features for object image classification
AU - Wang, Gang
AU - Hoiem, Derek
AU - Forsyth, David
PY - 2009/1/1
Y1 - 2009/1/1
N2 - We introduce a text-based image feature and demon- strate that it consistently improves performance on hard object classification problems. The feature is built using an auxiliary dataset of images annotated with tags, down- loaded from the internet. We do not inspect or correct the tags and expect that they are noisy. We obtain the text fea- ture of an unannotated image from the tags of its k-nearest neighbors in this auxiliary collection. A visual classifier presented with an object viewed un- der novel circumstances (say, a new viewing direction) must rely on its visual examples. Our text feature may not change, because the auxiliary dataset likely contains a similar pic- ture. While the tags associated with images are noisy, they are more stable when appearance changes. We test the performance of this feature using PAS- CAL VOC 2006 and 2007 datasets. Our feature performs well, consistently improves the performance of visual ob- ject classifiers, and is particularly effective when the train- ing dataset is small.
AB - We introduce a text-based image feature and demon- strate that it consistently improves performance on hard object classification problems. The feature is built using an auxiliary dataset of images annotated with tags, down- loaded from the internet. We do not inspect or correct the tags and expect that they are noisy. We obtain the text fea- ture of an unannotated image from the tags of its k-nearest neighbors in this auxiliary collection. A visual classifier presented with an object viewed un- der novel circumstances (say, a new viewing direction) must rely on its visual examples. Our text feature may not change, because the auxiliary dataset likely contains a similar pic- ture. While the tags associated with images are noisy, they are more stable when appearance changes. We test the performance of this feature using PAS- CAL VOC 2006 and 2007 datasets. Our feature performs well, consistently improves the performance of visual ob- ject classifiers, and is particularly effective when the train- ing dataset is small.
UR - http://www.scopus.com/inward/record.url?scp=70450207253&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=70450207253&partnerID=8YFLogxK
U2 - 10.1109/CVPRW.2009.5206816
DO - 10.1109/CVPRW.2009.5206816
M3 - Conference contribution
AN - SCOPUS:70450207253
SN - 9781424439935
T3 - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
SP - 1367
EP - 1374
BT - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
PB - IEEE Computer Society
T2 - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Y2 - 20 June 2009 through 25 June 2009
ER -