Prompt Augmentation for Self-supervised Text-guided Image Manipulation

TL;DR

This paper introduces prompt augmentation and a contrastive loss to improve text-guided image editing, enhancing localised manipulation and context preservation in diffusion models, with demonstrated superior results on benchmark datasets.

Contribution

It proposes a novel prompt augmentation technique combined with a contrastive and soft contrastive loss for more effective and localized text-guided image editing.

Findings

Improved image editing quality over state-of-the-art methods.

Enhanced localised manipulation and context preservation.

Effective on public datasets and generated images.

Abstract

Text-guided image editing finds applications in various creative and practical fields. While recent studies in image generation have advanced the field, they often struggle with the dual challenges of coherent image transformation and context preservation. In response, our work introduces prompt augmentation, a method amplifying a single input prompt into several target prompts, strengthening textual context and enabling localised image editing. Specifically, we use the augmented prompts to delineate the intended manipulation area. We propose a Contrastive Loss tailored to driving effective image editing by displacing edited areas and drawing preserved regions closer. Acknowledging the continuous nature of image manipulations, we further refine our approach by incorporating the similarity concept, creating a Soft Contrastive Loss. The new losses are incorporated to the diffusion model, demonstrating improved or competitive image editing results on public datasets and generated images over state-of-the-art approaches.