SemLayer: Semantic-aware Generative Segmentation and Layer Construction for Abstract Icons

TL;DR

SemLayer is a novel pipeline that restores semantic layers in flattened icons, enabling editing and restyling by generating separable components, reconstructing occluded parts, and assembling layered vector graphics.

Contribution

We introduce SemLayer, the first method for semantic layer construction in flattened icons, combining visual generation and shape completion for editable layered vector graphics.

Findings

Effective semantic layer reconstruction demonstrated through qualitative comparisons.

Quantitative evaluations show improved accuracy in recovering original icon structures.

Enables new editing workflows for flattened vector graphics.

Abstract

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/