TY - GEN
T1 - Phase Consistent Ecological Domain Adaptation
AU - Yang, Yanchao
AU - Alzahrani, Majed A.
AU - Sundaramoorthi, Ganesh
AU - Soatto, Stefano
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: Research supported by ARO W911NF-17-1-0304 and ONR N00014-19-1-2066. Dong Lao is supported by KAUST through the VCC Center Competitive Funding
PY - 2020
Y1 - 2020
N2 - We introduce two criteria to regularize the optimization involved in learning a classifier in a domain where no annotated data are available, leveraging annotated data in a different domain, a problem known as unsupervised domain adaptation. We focus on the task of semantic segmentation, where annotated synthetic data are aplenty, but annotating real data is laborious. The first criterion, inspired by visual psychophysics, is that the map between the two image domains be phase-preserving. This restricts the set of possible learned maps, while enabling enough flexibility to transfer semantic information. The second criterion aims to leverage ecological statistics, or regularities in the scene which are manifest in any image of it, regardless of the characteristics of the illuminant or the imaging sensor. It is implemented using a deep neural network that scores the likelihood of each possible segmentation given a single un-annotated image. Incorporating these two priors in a standard domain adaptation framework improves performance across the board in the most common unsupervised domain adaptation benchmarks for semantic segmentation.
AB - We introduce two criteria to regularize the optimization involved in learning a classifier in a domain where no annotated data are available, leveraging annotated data in a different domain, a problem known as unsupervised domain adaptation. We focus on the task of semantic segmentation, where annotated synthetic data are aplenty, but annotating real data is laborious. The first criterion, inspired by visual psychophysics, is that the map between the two image domains be phase-preserving. This restricts the set of possible learned maps, while enabling enough flexibility to transfer semantic information. The second criterion aims to leverage ecological statistics, or regularities in the scene which are manifest in any image of it, regardless of the characteristics of the illuminant or the imaging sensor. It is implemented using a deep neural network that scores the likelihood of each possible segmentation given a single un-annotated image. Incorporating these two priors in a standard domain adaptation framework improves performance across the board in the most common unsupervised domain adaptation benchmarks for semantic segmentation.
UR - http://hdl.handle.net/10754/662571
UR - https://ieeexplore.ieee.org/document/9157388/
U2 - 10.1109/CVPR42600.2020.00903
DO - 10.1109/CVPR42600.2020.00903
M3 - Conference contribution
SN - 978-1-7281-7169-2
BT - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
PB - IEEE
ER -