LoRAShop: Training-Free Multi-Concept Image Generation and Editing with Rectified Flow Transformers

TL;DR

LoRAShop is a training-free framework that enables multi-concept image editing by leveraging feature interaction patterns in diffusion transformers, allowing seamless integration of multiple subjects or styles.

Contribution

It introduces a novel method for multi-concept editing using disentangled masks and LoRA weights without retraining, enhancing identity preservation and editing flexibility.

Findings

Outperforms baselines in identity preservation.

Enables seamless multi-concept editing without retraining.

Facilitates rapid creative image manipulation.

Abstract

We introduce LoRAShop, the first framework for multi-concept image editing with LoRA models. LoRAShop builds on a key observation about the feature interaction patterns inside Flux-style diffusion transformers: concept-specific transformer features activate spatially coherent regions early in the denoising process. We harness this observation to derive a disentangled latent mask for each concept in a prior forward pass and blend the corresponding LoRA weights only within regions bounding the concepts to be personalized. The resulting edits seamlessly integrate multiple subjects or styles into the original scene while preserving global context, lighting, and fine details. Our experiments demonstrate that LoRAShop delivers better identity preservation compared to baselines. By eliminating retraining and external constraints, LoRAShop turns personalized diffusion models into a practical `photoshop-with-LoRAs' tool and opens new avenues for compositional visual storytelling and rapid creative iteration.