# Ranking CGANs: Subjective Control over Semantic Image Attributes

arXiv: https://arxiv.org/abs/1804.04082 · HTML: https://mlsanity.com/paper/1804.04082/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Apr 2018.

## Built on by (most cited first, 9 of 9)

- [Batch Active Learning at Scale](https://mlsanity.com/paper/2107.14263/) (2021)
- [A comprehensive survey on deep active learning in medical image analysis](https://mlsanity.com/paper/2310.14230/) (2023)
- [RankSRGAN: Super Resolution Generative Adversarial Networks with Learning to Rank](https://mlsanity.com/paper/2107.09427/) (2021)
- [Deep Active Learning for Computer Vision: Past and Future](https://mlsanity.com/paper/2211.14819/) (2022)
- [A Survey on Active Deep Learning: From Model-driven to Data-driven](https://mlsanity.com/paper/2101.09933/) (2021)
- [Semi-Supervised Variational Adversarial Active Learning via Learning to Rank and Agreement-Based Pseudo Labeling](https://mlsanity.com/paper/2408.12774/) (2024)
- [TRIP: Refining Image-to-Image Translation via Rival Preferences](https://mlsanity.com/paper/2111.13411/) (2021)
- [Subjective Face Transform using Human First Impressions](https://mlsanity.com/paper/2309.15381/) (2023)
- [ADROIT: A Self-Supervised Framework for Learning Robust Representations for Active Learning](https://mlsanity.com/paper/2503.07506/) (2025)

## Abstract (arXiv)

In this paper, we investigate the use of generative adversarial networks in the task of image generation according to subjective measures of semantic attributes. Unlike the standard (CGAN) that generates images from discrete categorical labels, our architecture handles both continuous and discrete scales. Given pairwise comparisons of images, our model, called RankCGAN, performs two tasks: it learns to rank images using a subjective measure; and it learns a generative model that can be controlled by that measure. RankCGAN associates each subjective measure of interest to a distinct dimension of some latent space. We perform experiments on UT-Zap50K, PubFig and OSR datasets and demonstrate that the model is expressive and diverse enough to conduct two-attribute exploration and image editing.