Realistic Saliency Guided Image Enhancement

 S. Mahdi H. MiangolehZoya BylinskiiEric KeeEli ShechtmanYağız Aksoy
Proc. CVPR, 2023
Realistic Saliency Guided Image Enhancement

(top) We develop a saliency-based image enhancement method that can be applied to multiple regions in the image to de-emphasize objects (steps 1, 2) or enhance subjects (steps 3, 4). (bottom) Our novel realism loss allows us to apply realistic edits to a wide variety of objects while state-of-the-art methods GazeShift and DeepSal may generate less realistic editing results.


Common editing operations performed by professional photographers include the cleanup operations: de-emphasizing distracting elements and enhancing subjects. These edits are challenging, requiring a delicate balance between manipulating the viewer's attention while maintaining photo realism. While recent approaches can boast successful examples of attention attenuation or amplification, most of them also suffer from frequent unrealistic edits. We propose a realism loss for saliency-guided image enhancement to maintain high realism across varying image types, while attenuating distractors and amplifying objects of interest. Evaluations with professional photographers confirm that we achieve the dual objective of realism and effectiveness, and outperform the recent approaches on their own datasets, while requiring a smaller memory footprint and runtime. We thus offer a viable solution for automating image enhancement and photo cleanup operations.



Training and Inference GitHub Repository




author={S. Mahdi H. Miangoleh and Zoya Bylinskii and Eric Kee and Eli Shechtman and Ya\u{g}{\i}z Aksoy},
title={Realistic Saliency Guided Image Enhancement},
journal={Proc. CVPR},

Related Publications

Chris Careaga, S. Mahdi H. Miangoleh, and Yağız Aksoy
SIGGRAPH Asia, 2023
Despite significant advancements in network-based image harmonization techniques, there still exists a domain disparity between typical training pairs and real-world composites encountered during inference. Most existing methods are trained to reverse global edits made on segmented image regions, which fail to accurately capture the lighting inconsistencies between the foreground and background found in composited images. In this work, we introduce a self-supervised illumination harmonization approach formulated in the intrinsic image domain. First, we estimate a simple global lighting model from mid-level vision representations to generate a rough shading for the foreground region. A network then refines this inferred shading to generate a harmonious re-shading that aligns with the background scene. In order to match the color appearance of the foreground and background, we utilize ideas from prior harmonization approaches to perform parameterized image edits in the albedo domain. To validate the effectiveness of our approach, we present results from challenging real-world composites and conduct a user study to objectively measure the enhanced realism achieved compared to state-of-the-art harmonization methods.
author={Chris Careaga and S. Mahdi H. Miangoleh and Ya\u{g}{\i}z Aksoy},
title={Intrinsic Harmonization for Illumination-Aware Compositing},
booktitle={Proc. SIGGRAPH Asia},