4.7 Article

Neural Modeling of Portrait Bas-Relief From a Single Photograph

期刊

出版社

IEEE COMPUTER SOC
DOI: 10.1109/TVCG.2022.3197354

关键词

Bas-relief modeling; depth reconstruction; image-to-depth translation; portrait bas-relief

向作者/读者索取更多资源

This paper presents an end-to-end neural solution for modeling portrait bas-relief from a single photograph. The authors address the challenge of lacking bas-relief data by proposing a semi-automatic pipeline to synthesize bas-relief samples. They train different network architectures on synthetic data and select the best-performing one through qualitative and quantitative comparisons. Experiments, comparisons, and evaluations by artists demonstrate the effectiveness and efficiency of the selected network.
In this paper, we present an end-to-end neural solution to model portrait bas-relief from a single photograph, which is cast as a problem of image-to-depth translation. The main challenge is the lack of bas-relief data for network training. To solve this problem, we propose a semi-automatic pipeline to synthesize bas-relief samples. The main idea is to first construct normal maps from photos, and then generate bas-relief samples by reconstructing pixel-wise depths. In total, our synthetic dataset contains 23 k pixel-wise photo/bas-relief pairs. Since the process of bas-relief synthesis requires a certain amount of user interactions, we propose end-to-end solutions with various network architectures, and train them on the synthetic data. We select the one that gave the best results through qualitative and quantitative comparisons. Experiments on numerous portrait photos, comparisons with state-of-the-art methods and evaluations by artists have proven the effectiveness and efficiency of the selected network.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据