SVG360: Multi-View SVG Generation with Geometric and Color Consistency from a Single SVG

2025-11-20

AI Summary

This work addresses the challenge of generating consistent multi-view SVGs from a single input view, specifically tackling cross-view geometric distortion and color inconsistency. The proposed method introduces an end-to-end framework grounded in 3D reconstruction, incorporating a cross-view spatial memory mechanism to establish part-level multi-view correspondences. It integrates an extended SAM2-based spatial memory bank, rasterization-assisted supervision, path merging, and structural optimization—enabling path and coloring consistency without retraining. Its key contribution is the first application of spatial memory mechanisms to multi-view vector graphics generation, uniquely balancing structural preservation with redundancy removal. Experiments demonstrate significant improvements over state-of-the-art methods in geometric fidelity, color consistency, and fine-detail retention, enabling high-quality vector asset creation and semantics-aware editing.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-instance/Multi-view Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Scalable Vector Graphics (SVGs) are central to modern design workflows, offering scaling without distortion and precise editability. However, for single object SVGs, generating multi-view consistent SVGs from a single-view input remains underexplored. We present a three stage framework that produces multi-view SVGs with geometric and color consistency from a single SVG input. First, the rasterized input is lifted to a 3D representation and rendered under target camera poses, producing multi-view images of the object. Next, we extend the temporal memory mechanism of Segment Anything 2 (SAM2) to the spatial domain, constructing a spatial memory bank that establishes part level correspondences across neighboring views, yielding cleaner and more consistent vector paths and color assignments without retraining. Finally, during the raster to vector conversion, we perform path consolidation and structural optimization to reduce redundancy while preserving boundaries and semantics. The resulting SVGs exhibit strong geometric and color consistency across views, significantly reduce redundant paths, and retain fine structural details. This work bridges generative modeling and structured vector representation, providing a scalable route to single input, object level multi-view SVG generation and supporting applications such as asset creation and semantic vector editing.

Problem

• Generating multi-view consistent SVGs from single-view input with geometric and color consistency
• Establishing part-level correspondences across views without retraining for cleaner vector paths
• Reducing redundant paths while preserving boundaries and semantics during vector conversion

Innovation

• Lifts single SVG to 3D representation for multi-view rendering
• Extends SAM2 temporal memory to spatial correspondence mechanism
• Performs path consolidation and structural optimization during vectorization