TY - GEN
T1 - A two level approach for scene recognition
AU - Lu, Le
AU - Toyama, Kentaro
AU - Hager, Gregory D.
PY - 2005
Y1 - 2005
N2 - Classifying pictures into one of several semantic categories is a classical image understanding problem. In this paper, we present a stratified approach to both binary (outdoor-indoor) and multiple category of scene classification. We first learn mixture models for 20 basic classes of local image content based on color and texture information. Once trained, these models are applied to a test image, and produce 20 probability density response maps (PDRM) indicating the likelihood that each image region was produced by each class. We then extract some very simple features from those PDRMs, and use them to train a bagged LDA classifier for 10 scene categories. For this process, no explicit region segmentation or spatial context model are computed. To test this classification system, we created a labeled database of 1500 photos taken under very different environment and lighting conditions, using different cameras, and from 43 persons over 5 years. The classification rate of outdoor-indoor classification is 93.8%. and the classification rate for 10 scene categories is 90.1%. As a byproduct, local image patches can be contextually labeled into the 20 basic material classes by using Loopy Belief Propagation [33] as an anisotropic filter on PDRMs, producing an image-level segmentation if desired.
AB - Classifying pictures into one of several semantic categories is a classical image understanding problem. In this paper, we present a stratified approach to both binary (outdoor-indoor) and multiple category of scene classification. We first learn mixture models for 20 basic classes of local image content based on color and texture information. Once trained, these models are applied to a test image, and produce 20 probability density response maps (PDRM) indicating the likelihood that each image region was produced by each class. We then extract some very simple features from those PDRMs, and use them to train a bagged LDA classifier for 10 scene categories. For this process, no explicit region segmentation or spatial context model are computed. To test this classification system, we created a labeled database of 1500 photos taken under very different environment and lighting conditions, using different cameras, and from 43 persons over 5 years. The classification rate of outdoor-indoor classification is 93.8%. and the classification rate for 10 scene categories is 90.1%. As a byproduct, local image patches can be contextually labeled into the 20 basic material classes by using Loopy Belief Propagation [33] as an anisotropic filter on PDRMs, producing an image-level segmentation if desired.
UR - http://www.scopus.com/inward/record.url?scp=33745119067&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33745119067&partnerID=8YFLogxK
U2 - 10.1109/cvpr.2005.51
DO - 10.1109/cvpr.2005.51
M3 - Conference contribution
AN - SCOPUS:33745119067
SN - 0769523722
SN - 9780769523729
T3 - Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
SP - 688
EP - 695
BT - Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
PB - IEEE Computer Society
T2 - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005
Y2 - 20 June 2005 through 25 June 2005
ER -