PortraVec: Image-Based Portrait Vectorization with Text-Guided Manipulation

TL;DR

PortraVec is a novel framework that converts portrait images into editable vector sketches with text-guided semantic control, improving structural accuracy and detail preservation.

Contribution

It introduces a two-stage image-guided vectorization method with attention-aware sampling and a text-guided editing module for enhanced portrait manipulation.

Findings

Achieves superior structural consistency compared to existing methods.

Maintains high visual fidelity and detail in vectorized portraits.

Enables local semantic editing with global consistency.

Abstract

While portrait sketch generation is a special task in sketch synthesis, most existing methods are pixel-based, limiting their interpretability and editability. With the rise of vector generation techniques, representing sketches using vector elements may provide more flexible manipulation. However, due to the overlapping nature of vector graphics and the coarse detail modeling, existing vectorization methods struggle to capture facial integrity and fine-grained details, and lack semantic control. To address these issues, we propose PortraVec, a framework for converting pixel-based portrait images into vector sketches with text control. Specifically, we propose a two-stage image-guided generation module using Attention-aware Offset Sampling to capture face structure while correcting detail deviations, and a text-guided manipulation module based on Region-based Parameter Freezing to enable local semantic editing while maintaining global consistency. Experiments show that PortraVec achieves superior structural consistency, visual fidelity, and semantic controllability compared to state-of-the-art methods.