Beyond Simple Edits: X-Planner for Complex Instruction-Based Image Editing
This addresses the problem of complex instruction-based image editing for users of diffusion models, representing a novel method for a known bottleneck rather than incremental.
The paper tackles the problem of diffusion-based image editing models struggling with complex, indirect instructions and poor identity preservation by introducing X-Planner, an MLLM-based planning system that decomposes instructions into sub-instructions and generates precise edit types and masks, achieving state-of-the-art results on benchmarks.
Recent diffusion-based image editing methods have significantly advanced text-guided tasks but often struggle to interpret complex, indirect instructions. Moreover, current models frequently suffer from poor identity preservation, unintended edits, or rely heavily on manual masks. To address these challenges, we introduce X-Planner, a Multimodal Large Language Model (MLLM)-based planning system that effectively bridges user intent with editing model capabilities. X-Planner employs chain-of-thought reasoning to systematically decompose complex instructions into simpler, clear sub-instructions. For each sub-instruction, X-Planner automatically generates precise edit types and segmentation masks, eliminating manual intervention and ensuring localized, identity-preserving edits. Additionally, we propose a novel automated pipeline for generating large-scale data to train X-Planner which achieves state-of-the-art results on both existing benchmarks and our newly introduced complex editing benchmark.