Frequency-Controlled Diffusion Model for Versatile Text-Guided Image-to-Image Translation

Abstract

Recently, large-scale text-to-image (T2I) diffusion models have emerged as a powerful tool for image-to-image translation (I2I), allowing open-domain image translation via user-provided text prompts. This paper proposes frequency-controlled diffusion model (FCDiffusion), an end-to-end diffusion-based framework that contributes a novel solution to text-guided I2I from a frequency-domain perspective. At the heart of our framework is a feature-space frequency-domain filtering module based on Discrete Cosine Transform, which filters the latent features of the source image in the DCT domain, yielding filtered image features bearing different DCT spectral bands as different control signals to the pre-trained Latent Diffusion Model. We reveal that control signals of different DCT spectral bands bridge the source image and the T2I generated image in different correlations (e.g., style, structure, layout, contour, etc.), and thus enable versatile I2I applications emphasizing different I2I correlations, including style-guided content creation, image semantic manipulation, image scene translation, and image style translation. Different from related approaches, FCDiffusion establishes a unified text-guided I2I framework suitable for diverse image translation tasks simply by switching among different frequency control branches at inference time. The effectiveness and superiority of our method for text-guided I2I are demonstrated with extensive experiments both qualitatively and quantitatively. Our project is publicly available at: https://xianggao1102.github.io/FCDiffusion/.

Proceedings of the 38th AAAI Conference on Artificial Intelligence (AAAI 2024)

References in corpus (13)

• Hierarchical Text-Conditional Image Generation with CLIP Latents
• Diffusion Models Beat GANs on Image Synthesis
• High-Resolution Image Synthesis with Latent Diffusion Models
• Prompt-to-Prompt Image Editing with Cross Attention Control
• Adding Conditional Control to Text-to-Image Diffusion Models
• Contrastive Learning for Unpaired Image-to-Image Translation
• Pretraining is All You Need for Image-to-Image Translation
• Diffusion-based Image Translation using Disentangled Style and Content Representation
• InstructPix2Pix: Learning to Follow Image Editing Instructions
• Text2LIVE: Text-Driven Layered Image and Video Editing
• RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and Generation
• Splicing ViT Features for Semantic Appearance Transfer
• Learning to Incorporate Texture Saliency Adaptive Attention to Image Cartoonization