Learning Continuous Implicit Representation for Near-Periodic Patterns
Learning Continuous Implicit Representation for Near-Periodic Patterns
复制标题
DOI：
10.48550/arxiv.2208.12278
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
B. Chen;Tiancheng Zhi;M. Hebert;S. Narasimhan
中科院分区：
文献类型：
--
作者：
B. Chen;Tiancheng Zhi;M. Hebert;S. Narasimhan
近周期图案（Near-Periodic Patterns，NPP）在人造场景中普遍存在，并且由具有由照明、缺陷或设计元素引起的外观差异的平铺图案组成。一个好的NPP表示是有用的许多应用，包括图像完成，分割，几何重映射。但是，表示NPP是具有挑战性的，因为它需要保持全局一致性（平铺图案布局），同时保留局部变化（外观差异）。使用大型数据集或单图像优化在一般场景上训练的方法很难满足这些约束，而显式建模周期性的方法对周期性检测错误不鲁棒。为了解决这些挑战，我们使用基于坐标的MLP和单个图像优化来学习神经隐式表示。我们设计了一个输入特征扭曲模块和一个基于一致性的补丁丢失模块来处理全局一致性和局部变化。为了进一步提高鲁棒性，我们引入了一个周期性建议模块来搜索和使用我们管道中的多个候选周期。我们证明了我们的方法的有效性，超过500个图像的建筑立面，中楣，壁纸，地面，蒙德里安模式的单一和多平面场景。
Near-Periodic Patterns (NPP) are ubiquitous in man-made scenes and are composed of tiled motifs with appearance differences caused by lighting, defects, or design elements. A good NPP representation is useful for many applications including image completion, segmentation, and geometric remapping. But representing NPP is challenging because it needs to maintain global consistency (tiled motifs layout) while preserving local variations (appearance differences). Methods trained on general scenes using a large dataset or single-image optimization struggle to satisfy these constraints, while methods that explicitly model periodicity are not robust to periodicity detection errors. To address these challenges, we learn a neural implicit representation using a coordinate-based MLP with single image optimization. We design an input feature warping module and a periodicity-guided patch loss to handle both global consistency and local variations. To further improve the robustness, we introduce a periodicity proposal module to search and use multiple candidate periodicities in our pipeline. We demonstrate the effectiveness of our method on more than 500 images of building facades, friezes, wallpapers, ground, and Mondrian patterns on single and multi-planar scenes.