PulseAugur
LIVE 14:46:24
research · [1 source] ·
0
research

PortraVec framework vectorizes portraits with text-guided manipulation

Researchers have developed PortraVec, a new framework designed to convert pixel-based portrait images into editable vector sketches. Unlike previous pixel-based methods that offer limited manipulation, PortraVec represents sketches using vector elements for greater flexibility. The system employs a two-stage generation module for capturing facial structure and a text-guided module for semantic editing, enabling local changes while preserving overall consistency. Experiments indicate PortraVec surpasses existing methods in structural integrity, visual quality, and semantic control. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Enables more flexible and semantically controllable editing of portrait images through vectorization.

RANK_REASON Academic paper introducing a new method for image vectorization with text-guided manipulation.

Read on arXiv cs.CV →

COVERAGE [1]

  1. arXiv cs.CV TIER_1 · Yiqi Liang, Ying Liu, Dandan Long, Ruihui Li ·

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

    arXiv:2410.04182v2 Announce Type: replace 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…