SemLayer：面向抽象图标的语义感知生成式分割与图层构建
SemLayer: Semantic-aware Generative Segmentation and Layer Construction for Abstract Icons
将扁平化矢量图标分解为语义层，为编辑、重着色和动画提供可操作的分层结构；其语义补全和遮挡关系推断方法可迁移到其他需要恢复遮挡结构的图像编辑任务。
原文依据 SemLayer first generates a chromatically differentiated representation in which distinct semantic components become visually separable.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论图形图标是现代设计工作流的基石，但它们通常以扁平的单路径或复合路径图形分发，原始的语义分层信息已丢失。这种语义分解的缺失阻碍了编辑、重着色和动画等下游任务。我们将此问题形式化为扁平化矢量艺术的语义图层构建，并引入 SemLayer，一个恢复可编辑分层结构的视觉生成赋能流程。给定一个抽象图标，SemLayer 首先生成一种色彩差异化的表示，其中不同的语义组件在视觉上变得可分离。为了恢复每个部分的完整几何形状（包括被遮挡区域），我们随后执行语义补全步骤，重建连贯的物体级形状。最后，恢复的部分被组装成具有推断遮挡关系的分层矢量表示。大量的定性比较和定量评估证明了 SemLayer 的有效性，使得先前无法应用于扁平化矢量图形的编辑工作流成为可能，并确立了语义图层重建作为一个实用且有价值的任务。项目页面：https://xxuhaiyang.github.io/SemLayer/
英文原文摘要 1248 字符
Graphic icons are a cornerstone of modern design workflows, yet they are often distributed as flattened single-path or compound-path graphics, where the original semantic layering is lost. This absence of semantic decomposition hinders downstream tasks such as editing, restyling, and animation. We formalize this problem as semantic layer construction for flattened vector art and introduce SemLayer, a visual generation empowered pipeline that restores editable layered structures. Given an abstract icon, SemLayer first generates a chromatically differentiated representation in which distinct semantic components become visually separable. To recover the complete geometry of each part, including occluded regions, we then perform a semantic completion step that reconstructs coherent object-level shapes. Finally, the recovered parts are assembled into a layered vector representation with inferred occlusion relationships. Extensive qualitative comparisons and quantitative evaluations demonstrate the effectiveness of SemLayer, enabling editing workflows previously inapplicable to flattened vector graphics and establishing semantic layer reconstruction as a practical and valuable task. Project page: https://xxuhaiyang.github.io/SemLayer/