In this paper, we tackle the problem of saliency-guided image manipulation for adjusting the saliency distribution over image regions. Conventional approaches ordinarily utilize explicit operations on altering the low-level features based on the selected saliency computation. However, it is difficult to generalize such methods for various saliency estimations. To address this issue, we propose a deep learning-based model that bridges between any differentiable saliency estimation methods and a neural network which applies image manipulation. Thus, the manipulation can be directly optimized in order to satisfy saliency-guidance. Extensive experiment results verify the capacity of our model in saliency-driven image editing and show favorable performance against numerous baselines.
Supplementary material (ZIP)