2023/10/11 by Hye-Seong Hong, Abhishek Kumar, Hong, Hye-Seong +3
Computer Science · Medicine · #Artificial intelligence #COVID-19 diagnosis using AI #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Discriminator #Domain Adaptation and Few-Shot Learning #Entropy (arrow of time) #FOS: Computer and information sciences #Machine learning #Pattern recognition (psychology) #Robustness (evolution) #Segmentation #Unsupervised learning
paper · pdf · doi:10.48550/arxiv.2310.07149
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2023/10/11 · openalex created_date 2023/10/13 · openalex updated_date 2026/07/28
The generalization capability of unsupervised domain adaptation can mitigate the need for extensive pixel-level annotations to train semantic segmentation networks by training models on synthetic data as a source with computer-generated annotations. Entropy-based adversarial networks are proposed to improve source domain prediction; however, they disregard significant external information, such as edges, which have the potential to identify and distinguish various objects within an image accurately. To address this issue, we introduce a novel approach to domain adaptation, leveraging the synergy of internal and external information within entropy-based adversarial networks. In this approach, we enrich the discriminator network with edge-predicted probability values within this innovative framework to enhance the clarity of class boundaries. Furthermore, we devised a probability-sharing network that integrates diverse information for more effective segmentation. Incorporating object edges addresses a pivotal aspect of unsupervised domain adaptation that has frequently been neglected in the past -- the precise delineation of object boundaries. Conventional unsupervised domain adaptation methods usually center around aligning feature distributions and may not explicitly model object boundaries. Our approach effectively bridges this gap by offering clear guidance on object boundaries, thereby elevating the quality of domain adaptation. Our approach undergoes rigorous evaluation on the established unsupervised domain adaptation benchmarks, specifically in adapting SYNTHIA → Cityscapes and SYNTHIA → Mapillary. Experimental results show that the proposed model attains better performance than state-of-the-art methods. The superior performance across different unsupervised domain adaptation scenarios highlights the versatility and robustness of the proposed method.