Skip to main content
placeholder image

BiMaL: Bijective Maximum Likelihood Approach to Domain Adaptation in Semantic Scene Segmentation

Conference Paper


Abstract


  • Semantic segmentation aims to predict pixel-level labels. It has become a popular task in various computer vision applications. While fully supervised segmentation methods have achieved high accuracy on large-scale vision datasets, they are unable to generalize on a new test environment or a new domain well. In this work, we first introduce a new Unaligned Domain Score to measure the efficiency of a learned model on a new target domain in unsupervised manner. Then, we present the new Bijective Maximum Likelihood (BiMaL) loss that is a generalized form of the Adversarial Entropy Minimization without any assumption about pixel independence. We have evaluated the proposed BiMaL on two domains. The proposed BiMaL approach consistently outperforms the SOTA methods on empirical experiments on “SYNTHIA to Cityscapes”, “GTA5 to Cityscapes”, and “SYNTHIA to Vistas”.

Publication Date


  • 2021

Citation


  • Truong, T. D., Duong, C. N., Le, N., Phung, S. L., Rainwater, C., & Luu, K. (2021). BiMaL: Bijective Maximum Likelihood Approach to Domain Adaptation in Semantic Scene Segmentation. In Proceedings of the IEEE International Conference on Computer Vision (pp. 8528-8537). doi:10.1109/ICCV48922.2021.00843

Scopus Eid


  • 2-s2.0-85121338600

Web Of Science Accession Number


Start Page


  • 8528

End Page


  • 8537

Abstract


  • Semantic segmentation aims to predict pixel-level labels. It has become a popular task in various computer vision applications. While fully supervised segmentation methods have achieved high accuracy on large-scale vision datasets, they are unable to generalize on a new test environment or a new domain well. In this work, we first introduce a new Unaligned Domain Score to measure the efficiency of a learned model on a new target domain in unsupervised manner. Then, we present the new Bijective Maximum Likelihood (BiMaL) loss that is a generalized form of the Adversarial Entropy Minimization without any assumption about pixel independence. We have evaluated the proposed BiMaL on two domains. The proposed BiMaL approach consistently outperforms the SOTA methods on empirical experiments on “SYNTHIA to Cityscapes”, “GTA5 to Cityscapes”, and “SYNTHIA to Vistas”.

Publication Date


  • 2021

Citation


  • Truong, T. D., Duong, C. N., Le, N., Phung, S. L., Rainwater, C., & Luu, K. (2021). BiMaL: Bijective Maximum Likelihood Approach to Domain Adaptation in Semantic Scene Segmentation. In Proceedings of the IEEE International Conference on Computer Vision (pp. 8528-8537). doi:10.1109/ICCV48922.2021.00843

Scopus Eid


  • 2-s2.0-85121338600

Web Of Science Accession Number


Start Page


  • 8528

End Page


  • 8537