Bokeh-Loss GAN: Multi-Stage Adversarial Training for Realistic Edge-Aware Bokeh
Abstract
In this paper, we tackle the problem of monocular bokeh synthesis, where we attempt to render a shallow depth of field image from a single all-in-focus image. Unlike in DSLR cameras, this effect can not be captured directly in mobile cameras due to the physical constraints of the mobile aperture. We thus propose a network-based approach that is capable of rendering realistic monocular bokeh from single image inputs. To do this, we introduce three new edge-aware Bokeh Losses based on a predicted monocular depth map, that sharpens the foreground edges while blurring the background. This model is then finetuned using an adversarial loss to generate a realistic Bokeh effect. Experimental results show that our approach is capable of generating a pleasing, natural Bokeh effect with sharp edges while handling complicated scenes.
Cite
Text
Lee et al. "Bokeh-Loss GAN: Multi-Stage Adversarial Training for Realistic Edge-Aware Bokeh." European Conference on Computer Vision Workshops, 2022. doi:10.1007/978-3-031-25063-7_39Markdown
[Lee et al. "Bokeh-Loss GAN: Multi-Stage Adversarial Training for Realistic Edge-Aware Bokeh." European Conference on Computer Vision Workshops, 2022.](https://mlanthology.org/eccvw/2022/lee2022eccvw-bokehloss/) doi:10.1007/978-3-031-25063-7_39BibTeX
@inproceedings{lee2022eccvw-bokehloss,
title = {{Bokeh-Loss GAN: Multi-Stage Adversarial Training for Realistic Edge-Aware Bokeh}},
author = {Lee, Brian and Lei, Fei and Chen, Huaijin G. and Baudron, Alexis},
booktitle = {European Conference on Computer Vision Workshops},
year = {2022},
pages = {619-634},
doi = {10.1007/978-3-031-25063-7_39},
url = {https://mlanthology.org/eccvw/2022/lee2022eccvw-bokehloss/}
}