Is Generative Modeling-based Stylization Necessary for Domain Adaptation in Regression Tasks?
This work addresses the computational inefficiency of generative modeling in domain adaptation for regression, offering a more practical solution for real-world applications.
The paper investigates whether generative modeling-based stylization is necessary for domain adaptation in regression tasks, finding that input-level alignment has little effect compared to classification, and proposes Implicit Stylization (ImSty), a non-parametric feature-level method that achieves consistent improvements over state-of-the-art without computational overhead.
Unsupervised domain adaptation (UDA) aims to bridge the gap between source and target domains in the absence of target domain labels using two main techniques: input-level alignment (such as generative modeling and stylization) and feature-level alignment (which matches the distribution of the feature maps, e.g. gradient reversal layers). Motivated from the success of generative modeling for image classification, stylization-based methods were recently proposed for regression tasks, such as pose estimation. However, use of input-level alignment via generative modeling and stylization incur additional overhead and computational complexity which limit their use in real-world DA tasks. To investigate the role of input-level alignment for DA, we ask the following question: Is generative modeling-based stylization necessary for visual domain adaptation in regression? Surprisingly, we find that input-alignment has little effect on regression tasks as compared to classification. Based on these insights, we develop a non-parametric feature-level domain alignment method -- Implicit Stylization (ImSty) -- which results in consistent improvements over SOTA regression task, without the need for computationally intensive stylization and generative modeling. Our work conducts a critical evaluation of the role of generative modeling and stylization, at a time when these are also gaining popularity for domain generalization.