# Emotional Semantics-Preserved and Feature-Aligned CycleGAN for Visual Emotion Adaptation

arXiv: https://arxiv.org/abs/2011.12470 · HTML: https://mlsanity.com/paper/2011.12470/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Nov 2020.

## Built on by (most cited first, 4 of 4)

- [Affective Image Content Analysis: Two Decades Review and New Perspectives](https://mlsanity.com/paper/2106.16125/) (2021)
- [Computational Emotion Analysis From Images: Recent Advances and Future Directions](https://mlsanity.com/paper/2103.10798/) (2021)
- [Bridging Visual Affective Gap: Borrowing Textual Knowledge by Learning from Noisy Image-Text Pairs](https://mlsanity.com/paper/2511.17103/) (2025)
- [Level of agreement between emotions generated by Artificial Intelligence and human evaluation: a methodological proposal](https://mlsanity.com/paper/2410.08332/) (2024)

## Abstract (arXiv)

Thanks to large-scale labeled training data, deep neural networks (DNNs) have obtained remarkable success in many vision and multimedia tasks. However, because of the presence of domain shift, the learned knowledge of the well-trained DNNs cannot be well generalized to new domains or datasets that have few labels. Unsupervised domain adaptation (UDA) studies the problem of transferring models trained on one labeled source domain to another unlabeled target domain. In this paper, we focus on UDA in visual emotion analysis for both emotion distribution learning and dominant emotion classification. Specifically, we design a novel end-to-end cycle-consistent adversarial model, termed CycleEmotionGAN++. First, we generate an adapted domain to align the source and target domains on the pixel-level by improving CycleGAN with a multi-scale structured cycle-consistency loss. During the image translation, we propose a dynamic emotional semantic consistency loss to preserve the emotion labels of the source images. Second, we train a transferable task classifier on the adapted domain with feature-level alignment between the adapted and target domains. We conduct extensive UDA experiments on the Flickr-LDL & Twitter-LDL datasets for distribution learning and ArtPhoto & FI datasets for emotion classification. The results demonstrate the significant improvements yielded by the proposed CycleEmotionGAN++ as compared to state-of-the-art UDA approaches.