Considerable progress has recently been made in leveraging CLIP (Contrastive Language-Image Pre-Training) models for text-guided image manipulation. However, all existing works rely on additional generative models to ensure the quality of results, because CLIP alone cannot provide enough guidance information for fine-scale pixel-level changes. In this paper, we introduce CLIPVG, a text-guided image manipulation framework using differentiable vector graphics, which is also the first CLIP-based general image manipulation framework that does not require any additional generative models. We demonstrate that CLIPVG can not only achieve state-of-art performance in both semantic correctness and synthesis quality, but also is flexible enough to support various applications far beyond the capability of all existing methods.
@article{arxiv.2212.02122,
title = {CLIPVG: Text-Guided Image Manipulation Using Differentiable Vector Graphics},
author = {Yiren Song and Xuning Shao and Kang Chen and Weidong Zhang and Minzhe Li and Zhongliang Jing},
journal= {arXiv preprint arXiv:2212.02122},
year = {2023}
}