DragText: Rethinking Text Embedding in Point-based Image Editing
This work addresses a specific bottleneck in diffusion-based image editing for users needing more accurate and flexible control, representing an incremental improvement.
The paper tackles the problem of text embedding remaining constant during point-based image editing in diffusion models, which causes a discrepancy with changing image embeddings, and proposes DragText to optimize text embedding alongside dragging, enhancing performance with minimal code integration.
Point-based image editing enables accurate and flexible control through content dragging. However, the role of text embedding during the editing process has not been thoroughly investigated. A significant aspect that remains unexplored is the interaction between text and image embeddings. During the progressive editing in a diffusion model, the text embedding remains constant. As the image embedding increasingly diverges from its initial state, the discrepancy between the image and text embeddings presents a significant challenge. In this study, we found that the text prompt significantly influences the dragging process, particularly in maintaining content integrity and achieving the desired manipulation. Upon these insights, we propose DragText, which optimizes text embedding in conjunction with the dragging process to pair with the modified image embedding. Simultaneously, we regularize the text optimization process to preserve the integrity of the original text prompt. Our approach can be seamlessly integrated with existing diffusion-based drag methods, enhancing performance with only a few lines of code.