AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea
This work addresses the need for more accurate and versatile image editing tools for users, though it is incremental as it builds on existing diffusion models with improved data and training methods.
The paper tackles the problem of instruction-based image editing models struggling with complex user instructions due to low-quality training data, by introducing AnyEdit, a dataset of 2.5 million high-quality editing pairs across 20 types and five domains, which boosts diffusion-based model performance in experiments.
Instruction-based image editing aims to modify specific image elements with natural language instructions. However, current models in this domain often struggle to accurately execute complex user instructions, as they are trained on low-quality data with limited editing types. We present AnyEdit, a comprehensive multi-modal instruction editing dataset, comprising 2.5 million high-quality editing pairs spanning over 20 editing types and five domains. We ensure the diversity and quality of the AnyEdit collection through three aspects: initial data diversity, adaptive editing process, and automated selection of editing results. Using the dataset, we further train a novel AnyEdit Stable Diffusion with task-aware routing and learnable task embedding for unified image editing. Comprehensive experiments on three benchmark datasets show that AnyEdit consistently boosts the performance of diffusion-based editing models. This presents prospects for developing instruction-driven image editing models that support human creativity.