TY - GEN
T1 - Tell me what you see and i will show you where it is
AU - Xu, Jia
AU - Schwing, Alexander G.
AU - Urtasun, Raquel
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2014/9/24
Y1 - 2014/9/24
N2 - We tackle the problem of weakly labeled semantic segmentation, where the only source of annotation are image tags encoding which classes are present in the scene. This is an extremely difficult problem as no pixel-wise labelings are available, not even at training time. In this paper, we show that this problem can be formalized as an instance of learning in a latent structured prediction framework, where the graphical model encodes the presence and absence of a class as well as the assignments of semantic labels to superpixels. As a consequence, we are able to leverage standard algorithms with good theoretical properties. We demonstrate the effectiveness of our approach using the challenging SIFT-flow dataset and show average per-class accuracy improvements of 7% over the state-of-the-art.
AB - We tackle the problem of weakly labeled semantic segmentation, where the only source of annotation are image tags encoding which classes are present in the scene. This is an extremely difficult problem as no pixel-wise labelings are available, not even at training time. In this paper, we show that this problem can be formalized as an instance of learning in a latent structured prediction framework, where the graphical model encodes the presence and absence of a class as well as the assignments of semantic labels to superpixels. As a consequence, we are able to leverage standard algorithms with good theoretical properties. We demonstrate the effectiveness of our approach using the challenging SIFT-flow dataset and show average per-class accuracy improvements of 7% over the state-of-the-art.
KW - Graphical Model
KW - Semantic Segmentation
KW - Structured Prediction
KW - Weakly Supervised Learning
UR - http://www.scopus.com/inward/record.url?scp=84911404516&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84911404516&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2014.408
DO - 10.1109/CVPR.2014.408
M3 - Conference contribution
AN - SCOPUS:84911404516
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 3190
EP - 3197
BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
PB - IEEE Computer Society
T2 - 27th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014
Y2 - 23 June 2014 through 28 June 2014
ER -