Borrowing from anything: A generalizable framework for reference-guided instance editing
This addresses the challenge of disentangling intrinsic and extrinsic attributes for instance editing, which is incremental as it builds on existing disentanglement methods.
The paper tackles the problem of semantic entanglement in reference-guided instance editing by proposing GENIE, a framework that achieves explicit disentanglement, and demonstrates state-of-the-art fidelity and robustness on the AnyInsertion dataset.
Reference-guided instance editing is fundamentally limited by semantic entanglement, where a reference's intrinsic appearance is intertwined with its extrinsic attributes. The key challenge lies in disentangling what information should be borrowed from the reference, and determining how to apply it appropriately to the target. To tackle this challenge, we propose GENIE, a Generalizable Instance Editing framework capable of achieving explicit disentanglement. GENIE first corrects spatial misalignments with a Spatial Alignment Module (SAM). Then, an Adaptive Residual Scaling Module (ARSM) learns what to borrow by amplifying salient intrinsic cues while suppressing extrinsic attributes, while a Progressive Attention Fusion (PAF) mechanism learns how to render this appearance onto the target, preserving its structure. Extensive experiments on the challenging AnyInsertion dataset demonstrate that GENIE achieves state-of-the-art fidelity and robustness, setting a new standard for disentanglement-based instance editing.