CLIPVG: Text-Guided Image Manipulation Using Differentiable Vector Graphics
This addresses the challenge of fine-scale pixel-level changes in image manipulation for users in computer vision and graphics, offering a more flexible and high-quality solution.
The paper tackles the problem of text-guided image manipulation by introducing CLIPVG, a framework that uses differentiable vector graphics to achieve state-of-the-art performance in semantic correctness and synthesis quality without requiring additional generative models.
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.