AmodalSVG: Amodal Image Vectorization via Semantic Layer Peeling

2026-04-12

AI Summary

Existing image vectorization methods process only visible pixels and disregard occlusion relationships, resulting in SVG outputs with ambiguous semantics, incomplete geometry, and limited editability. This work proposes a two-stage framework: first, a vision-language model–guided Semantic Layer Peeling (SLP) decouples and completes occluded objects in the raster domain; second, an error-budget–driven Adaptive Layered Vectorization (ALV) converts each semantic layer independently into vector graphics. The approach achieves, for the first time, complete geometric reconstruction and semantically layered vectorization of occluded objects in natural images. It produces SVGs with high visual fidelity, full structural integrity, and distinct semantic layers, enabling object-level vector editing and overcoming fundamental limitations of conventional vectorization techniques.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Representation Learning
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

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.

Problem

• amodal image vectorization
• semantic entanglement
• geometric incompleteness
• occlusion
• SVG editability

Innovation

• amodal vectorization
• semantic layer peeling
• adaptive layered vectorization
• occlusion completion
• editable SVG