Self-Play Reinforcement Learning for Fast Image Retargeting
This work addresses a practical bottleneck for users needing fast image resizing, though it is incremental as it builds on an existing multi-operator approach.
The paper tackles the problem of slow processing time in multi-operator image retargeting by proposing a reinforcement learning agent with self-play rewards and dynamic loss weights, achieving a three orders of magnitude speedup while maintaining the same quality as the best-performing method.
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.