A Scalable Vector Graphics Latent Space

2026-08-22

AI Summary

该研究解决了矢量图形缺乏连续、密集和可逆潜在空间的问题，通过引入基于Transformer的自编码器SLS，学习SVG路径的紧凑表示，提高效率和通用性。

Abstract

Scalable Vector Graphics are a fundamental medium for resolution-independent visual content, yet the deep learning community lacks a continuous, dense, and invertible latent space for vector representations, the kind of foundational building block that Variational Autoencoders and their descendants have long provided for raster images. We introduce SLS (SVG Latent Space), a Transformer-based autoencoder that learns compact dense representations of individual SVG paths, the atomic visual elements from which any SVG image can be composed. By modeling SVG commands, coordinate data, and visual properties within a unified BPE-based token vocabulary, SLS learns fixed-size latent representations that jointly capture structure and appearance, and can be decoded back into valid, style-consistent SVG paths with high fidelity. The resulting embedding space is robust, invertible, and structured: embeddings lie on a unit hypersphere, enabling efficient similarity search, composition, and downstream conditioning through simple vector-space operations. Finally, we demonstrate that SLS generalizes across diverse tasks reducing their FLOPs by over 150 times compared to token-based approaches, and establishing a general-purpose latent foundation for vector graphics research.

Problem

• Scalable Vector Graphics
• latent space
• vector representations
• autoencoder
• SVG paths

Innovation

• Scalable Vector Graphics
• Transformer-based autoencoder
• BPE-based token vocabulary
• unit hypersphere
• FLOPs reduction