A User-Friendly Framework for Generating Model-Preferred Prompts in Text-to-Image Synthesis

Abstract

Well-designed prompts have demonstrated the potential to guide text-to-image models in generating amazing images. Although existing prompt engineering methods can provide high-level guidance, it is challenging for novice users to achieve the desired results by manually entering prompts due to a discrepancy between novice-user-input prompts and the model-preferred prompts. To bridge the distribution gap between user input behavior and model training datasets, we first construct a novel Coarse-Fine Granularity Prompts dataset (CFP) and propose a novel User-Friendly Fine-Grained Text Generation framework (UF-FGTG) for automated prompt optimization. For CFP, we construct a novel dataset for text-to-image tasks that combines coarse and fine-grained prompts to facilitate the development of automated prompt generation methods. For UF-FGTG, we propose a novel framework that automatically translates user-input prompts into model-preferred prompts. Specifically, we propose a prompt refiner that continually rewrites prompts to empower users to select results that align with their unique needs. Meanwhile, we integrate image-related loss functions from the text-to-image model into the training process of text generation to generate model-preferred prompts. Additionally, we propose an adaptive feature extraction module to ensure diversity in the generated results. Experiments demonstrate that our approach is capable of generating more visually appealing and diverse images than previous state-of-the-art methods, achieving an average improvement of 5% across six quality and aesthetic metrics.

Accepted by The 38th Annual AAAI Conference on Artificial Intelligence (AAAI 2024)

References in corpus (24)

• Decoupled Weight Decay Regularization
• Learning to Prompt for Vision-Language Models
• Hierarchical Text-Conditional Image Generation with CLIP Latents
• Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
• Scaling Instruction-Finetuned Language Models
• LAION-5B: An open large-scale dataset for training next generation image-text models
• GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models
• Blind Image Quality Assessment Using A Deep Bilinear Convolutional Neural Network
• High-Resolution Image Synthesis with Latent Diffusion Models
• Reproducible scaling laws for contrastive language-image learning
• Elucidating the Design Space of Diffusion-Based Generative Models
• A Taxonomy of Prompt Modifiers for Text-To-Image Generation
• Large-scale Text-to-Image Generation Models for Visual Artists' Creative Works
• RePrompt: Automatic Prompt Editing to Refine AI-Generative Art Towards Precise Expressions
• Best Prompts for Text-to-Image Models and How to Find Them
• StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesis
• Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery
• Optimizing Prompts for Text-to-Image Generation
• Design Guidelines for Prompt Engineering Text-to-Image Generative Models
• Human Preference Score: Better Aligning Text-to-Image Models with Human Preference
• Opal: Multimodal Image Generation for News Illustration
• Iterative Prompt Learning for Unsupervised Backlit Image Enhancement
• I Spy a Metaphor: Large Language Models and Diffusion Models Co-Create Visual Metaphors
• Collaborative Generative AI: Integrating GPT-k for Efficient Editing in Text-to-Image Generation