LayerTracer：透過擴散Transformer實現認知對齊的分層SVG合成

LayerTracer: Cognitive-Aligned Layered SVG Synthesis via Diffusion Transformer

February 3, 2025

作者: Yiren Song, Danze Chen, Mike Zheng Shou

cs.AI

摘要

生成與認知相符的分層SVG仍然具有挑戰性，因為現有方法傾向於產生過於簡化的單層輸出或因優化而導致形狀冗餘。我們提出LayerTracer，一個基於擴散Transformer的框架，通過從一個新穎的連續設計操作數據集中學習設計師的分層SVG創建過程來彌合這一差距。我們的方法分為兩個階段：首先，一個文本條件的DiT生成多階段光柵化建構藍圖，模擬人類設計工作流程。其次，逐層矢量化與路徑去重產生乾淨、可編輯的SVG。對於圖像矢量化，我們引入了一種有條件的擴散機制，將參考圖像編碼為潛在令牌，引導分層重構同時保持結構完整性。廣泛的實驗證明LayerTracer在生成質量和可編輯性方面優於基於優化和神經的基線，有效地使AI生成的矢量與專業設計認知相一致。

English

Generating cognitive-aligned layered SVGs remains challenging due to existing methods' tendencies toward either oversimplified single-layer outputs or optimization-induced shape redundancies. We propose LayerTracer, a diffusion transformer based framework that bridges this gap by learning designers' layered SVG creation processes from a novel dataset of sequential design operations. Our approach operates in two phases: First, a text-conditioned DiT generates multi-phase rasterized construction blueprints that simulate human design workflows. Second, layer-wise vectorization with path deduplication produces clean, editable SVGs. For image vectorization, we introduce a conditional diffusion mechanism that encodes reference images into latent tokens, guiding hierarchical reconstruction while preserving structural integrity. Extensive experiments demonstrate LayerTracer's superior performance against optimization-based and neural baselines in both generation quality and editability, effectively aligning AI-generated vectors with professional design cognition.