Inspired by BalaGAN, an unsupervised class-to-class translation model based on conditional contrastive learning is proposed to capture the classes-related shared latent space rather than domains-related shared latent space for domain variations.
This paper proposes an unsupervised class-to-class translation model based on conditional contrastive learning to tackle the domain variations problem.
This paper first performs unsupervised semantic clustering for each domain to divide them into multiple classes and then leverages the classification features ...
A truly unsupervised image-to-image translation model (TUNIT) that simultaneously learns to separate image domains and translates input images into the ...
For one varying domain, samples vary significantly in shape and size and have no domain labels. This paper proposes an unsupervised class-to-class translation ...
Unsupervised Class-to-Class Translation for domain variations. June 2023 ... In this paper, we propose a Simplified Unsupervised Image Translation (SUIT) model ...
Jan 17, 2024 · In this paper we present a method to solve the unsupervised multiple domain translation task using only a Variational Autoencoder (VAE) based model.
Feb 14, 2023 · In this paper we report a method that allows us to discover how patterns in disparate domains can be reversibly related using a computationally rigorous ...
Jul 1, 2022 · Unsupervised Domain Adaptation expects large amounts of target data to be effective, and this is emphasized even more when using deep models.
Article,. Unsupervised class-to-class translation for domain variations. Z. Cao, W. Wang, ...