CVApr 13

AmodalSVG: Amodal Image Vectorization via Semantic Layer Peeling

arXiv:2604.1094095.52 citationsh-index: 11
AI Analysis

For users of vector graphics editing, this work enables direct object-level editing in vector domain, a capability not supported by existing vectorization methods.

AmodalSVG produces semantically organized and geometrically complete SVG representations from natural images by reconstructing full object geometries including occluded regions, enabling object-level editing in the vector domain. It significantly outperforms prior methods in visual fidelity.

We introduce AmodalSVG, a new framework for amodal image vectorization that produces semantically organized and geometrically complete SVG representations from natural images. Existing vectorization methods operate under a modal paradigm: tracing only visible pixels and disregarding occlusion. Consequently, the resulting SVGs are semantically entangled and geometrically incomplete, limiting SVG's structural editability. In contrast, AmodalSVG reconstructs full object geometries, including occluded regions, into independent, editable vector layers. To achieve this, AmodalSVG reformulates image vectorization as a two-stage framework, performing semantic decoupling and completion in the raster domain to produce amodally complete semantic layers, which are then independently vectorized. In the first stage, we introduce Semantic Layer Peeling (SLP), a VLM-guided strategy that progressively decomposes an image into semantically coherent layers. By hybrid inpainting, SLP recovers complete object appearances under occlusions, enabling explicit semantic decoupling. To vectorize these layers efficiently, we propose Adaptive Layered Vectorization (ALV), which dynamically modulates the primitive budget via an error-budget-driven adjustment mechanism. Extensive experiments demonstrate that AmodalSVG significantly outperforms prior methods in visual fidelity. Moreover, the resulting amodal layers enable object-level editing directly in the vector domain, capabilities not supported by existing vectorization approaches. Code will be released upon acceptance.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes