Interaction-Consistent Object Removal via MLLM-Based Reasoning
This addresses the issue of semantic inconsistency in image editing for users needing realistic object removal, though it is incremental as it builds on existing MLLM and image editing methods.
The paper tackles the problem of object removal in images leaving behind interaction evidence, formalizing it as Interaction-Consistent Object Removal (ICOR) and proposing REORM, a framework using multimodal large language models to infer and remove associated elements, which outperforms state-of-the-art systems on the ICOREval benchmark.
Image-based object removal often erases only the named target, leaving behind interaction evidence that renders the result semantically inconsistent. We formalize this problem as Interaction-Consistent Object Removal (ICOR), which requires removing not only the target object but also associated interaction elements, such as lighting-dependent effects, physically connected objects, targetproduced elements, and contextually linked objects. To address this task, we propose Reasoning-Enhanced Object Removal with MLLM (REORM), a reasoningenhanced object removal framework that leverages multimodal large language models to infer which elements must be jointly removed. REORM features a modular design that integrates MLLM-driven analysis, mask-guided removal, and a self-correction mechanism, along with a local-deployment variant that supports accurate editing under limited resources. To support evaluation, we introduce ICOREval, a benchmark consisting of instruction-driven removals with rich interaction dependencies. On ICOREval, REORM outperforms state-of-the-art image editing systems, demonstrating its effectiveness in producing interactionconsistent results.