Beyond Sliders: Mastering the Art of Diffusion-based Image Manipulation

2025/09/14 by Yufei Tang, Tang, Yufei, Daiheng Gao +9
Arts and Humanities · Computer Science · Engineering · #Advanced Numerical Analysis Techniques #Adversarial system #Art, Technology, and Culture #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image (mathematics) #Image manipulation #Image quality #Morphing #Music Technology and Sound Studies #Realism #Rendering (computer graphics) #Robustness (evolution)

paper · pdf · doi:10.48550/arxiv.2509.11213

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2025/09/14 · openalex created_date 2025/10/12 · openalex updated_date 2026/08/05

Abstract

In the realm of image generation, the quest for realism and customization has never been more pressing. While existing methods like concept sliders have made strides, they often falter when it comes to no-AIGC images, particularly images captured in real world settings. To bridge this gap, we introduce Beyond Sliders, an innovative framework that integrates GANs and diffusion models to facilitate sophisticated image manipulation across diverse image categories. Improved upon concept sliders, our method refines the image through fine grained guidance both textual and visual in an adversarial manner, leading to a marked enhancement in image quality and realism. Extensive experimental validation confirms the robustness and versatility of Beyond Sliders across a spectrum of applications.

Citations

• SOEDiff: Efficient Distillation for Small Object Editing
• Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey
• Scaling Rectified Flow Transformers for High-Resolution Image Synthesis
• SCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation
• Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Models
• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis
• ReVersion: Diffusion-Based Relation Inversion from Images
• Understanding Diffusion Models: A Unified Perspective
• Compositional Visual Generation with Composable Diffusion Models
• High-Resolution Image Synthesis with Latent Diffusion Models
• High-Resolution Image Synthesis with Latent Diffusion Models
• Generative Adversarial Networks
• Alias-Free Generative Adversarial Networks
• LoRA: Low-Rank Adaptation of Large Language Models
• Diffusion Models Beat GANs on Image Synthesis
• Learning Transferable Visual Models From Natural Language Supervision
• Denoising Diffusion Probabilistic Models
• Analyzing and Improving the Image Quality of StyleGAN
• Generative Modeling by Estimating Gradients of the Data Distribution
• A Style-Based Generator Architecture for Generative Adversarial Networks
• The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
• In Defense of the Triplet Loss for Person Re-Identification
• Perceptual Losses for Real-Time Style Transfer and Super-Resolution
• Deep metric learning using Triplet network
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Backpropagation Applied to Handwritten Zip Code Recognition

Related

• Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Models
• DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulation
• All-in-One Slider for Attribute Manipulation in Diffusion Models
• DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing
• Prompt Sliders for Fine-Grained Control, Editing and Erasing of Concepts in Diffusion Models