# Self-Play Reinforcement Learning for Fast Image Retargeting

arXiv: https://arxiv.org/abs/2010.00909 · HTML: https://mlsanity.com/paper/2010.00909/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Oct 2020.

## Built on by (most cited first, 7 of 7)

- [A Survey on Self-play Methods in Reinforcement Learning](https://mlsanity.com/paper/2408.01072/) (2024)
- [Retargeting Visual Data with Deformation Fields](https://mlsanity.com/paper/2311.13297/) (2023)
- [Fast Hybrid Image Retargeting](https://mlsanity.com/paper/2203.13595/) (2022)
- [Prune and Repaint: Content-Aware Image Retargeting for any Ratio](https://mlsanity.com/paper/2410.22865/) (2024)
- [HALO: Human-Aligned End-to-end Image Retargeting with Layered Transformations](https://mlsanity.com/paper/2504.03026/) (2025)
- [Supervised Deep Learning for Content-Aware Image Retargeting with Fourier Convolutions](https://mlsanity.com/paper/2306.07383/) (2023)
- [Competing for pixels: a self-play algorithm for weakly-supervised segmentation](https://mlsanity.com/paper/2405.16628/) (2024)

## Abstract (arXiv)

In this study, we address image retargeting, which is a task that adjusts input images to arbitrary sizes. In one of the best-performing methods called MULTIOP, multiple retargeting operators were combined and retargeted images at each stage were generated to find the optimal sequence of operators that minimized the distance between original and retargeted images. The limitation of this method is in its tremendous processing time, which severely prohibits its practical use. Therefore, the purpose of this study is to find the optimal combination of operators within a reasonable processing time; we propose a method of predicting the optimal operator for each step using a reinforcement learning agent. The technical contributions of this study are as follows. Firstly, we propose a reward based on self-play, which will be insensitive to the large variance in the content-dependent distance measured in MULTIOP. Secondly, we propose to dynamically change the loss weight for each action to prevent the algorithm from falling into a local optimum and from choosing only the most frequently used operator in its training. Our experiments showed that we achieved multi-operator image retargeting with less processing time by three orders of magnitude and the same quality as the original multi-operator-based method, which was the best-performing algorithm in retargeting tasks.