TY - GEN
T1 - SuperParsing
T2 - 11th European Conference on Computer Vision, ECCV 2010
AU - Tighe, Joseph
AU - Lazebnik, Svetlana
PY - 2010
Y1 - 2010
N2 - This paper presents a simple and effective nonparametric approach to the problem of image parsing, or labeling image regions (in our case, superpixels produced by bottom-up segmentation) with their categories. This approach requires no training, and it can easily scale to datasets with tens of thousands of images and hundreds of labels. It works by scene-level matching with global image descriptors, followed by superpixel-level matching with local features and efficient Markov random field (MRF) optimization for incorporating neighborhood context. Our MRF setup can also compute a simultaneous labeling of image regions into semantic classes (e.g., tree, building, car) and geometric classes (sky, vertical, ground). Our system outperforms the state-of-the-art nonparametric method based on SIFT Flow on a dataset of 2,688 images and 33 labels. In addition, we report per-pixel rates on a larger dataset of 15,150 images and 170 labels. To our knowledge, this is the first complete evaluation of image parsing on a dataset of this size, and it establishes a new benchmark for the problem.
AB - This paper presents a simple and effective nonparametric approach to the problem of image parsing, or labeling image regions (in our case, superpixels produced by bottom-up segmentation) with their categories. This approach requires no training, and it can easily scale to datasets with tens of thousands of images and hundreds of labels. It works by scene-level matching with global image descriptors, followed by superpixel-level matching with local features and efficient Markov random field (MRF) optimization for incorporating neighborhood context. Our MRF setup can also compute a simultaneous labeling of image regions into semantic classes (e.g., tree, building, car) and geometric classes (sky, vertical, ground). Our system outperforms the state-of-the-art nonparametric method based on SIFT Flow on a dataset of 2,688 images and 33 labels. In addition, we report per-pixel rates on a larger dataset of 15,150 images and 170 labels. To our knowledge, this is the first complete evaluation of image parsing on a dataset of this size, and it establishes a new benchmark for the problem.
KW - image parsing
KW - image segmentation
KW - scene understanding
UR - http://www.scopus.com/inward/record.url?scp=78149311874&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=78149311874&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-15555-0_26
DO - 10.1007/978-3-642-15555-0_26
M3 - Conference contribution
AN - SCOPUS:78149311874
SN - 3642155545
SN - 9783642155543
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 352
EP - 365
BT - Computer Vision, ECCV 2010 - 11th European Conference on Computer Vision, Proceedings
PB - Springer
Y2 - 10 September 2010 through 11 September 2010
ER -