Dynamic Frequency Modulation for Controllable Text-driven Image Generation

TL;DR

This paper introduces a frequency-based, training-free modulation technique for text-driven image generation that enhances semantic editing control while maintaining structural consistency, outperforming existing methods.

Contribution

It proposes a novel frequency modulation approach that directly manipulates latent variables, avoiding empirical feature selection and improving semantic editing stability.

Findings

Outperforms state-of-the-art methods in balancing structure preservation and semantic editing.

Effectively manipulates frequency spectrum to control image generation details.

Demonstrates robustness across various image editing scenarios.

Abstract

The success of text-guided diffusion models has established a new image generation paradigm driven by the iterative refinement of text prompts. However, modifying the original text prompt to achieve the expected semantic adjustments often results in unintended global structure changes that disrupt user intent. Existing methods rely on empirical feature map selection for intervention, whose performance heavily depends on appropriate selection, leading to suboptimal stability. This paper tries to solve the aforementioned problem from a frequency perspective and analyzes the impact of the frequency spectrum of noisy latent variables on the hierarchical emergence of the structure framework and fine-grained textures during the generation process. We find that lower-frequency components are primarily responsible for establishing the structure framework in the early generation stage. Their influence diminishes over time, giving way to higher-frequency components that synthesize fine-grained textures. In light of this, we propose a training-free frequency modulation method utilizing a frequency-dependent weighting function with dynamic decay. This method maintains the structure framework consistency while permitting targeted semantic modifications. By directly manipulating the noisy latent variable, the proposed method avoids the empirical selection of internal feature maps. Extensive experiments demonstrate that the proposed method significantly outperforms current state-of-the-art methods, achieving an effective balance between preserving structure and enabling semantic updates.