RoPEMover: Depth-Aware Object Relocation via Positional Embeddings

2026-06-25

AI Summary

Existing methods for single-image object relocation struggle to preserve geometric and scene consistency, particularly in handling occlusions, inpainting newly exposed regions, and updating shadows and reflections. This work proposes a geometry-aware object relocation approach based on a diffusion Transformer, which for the first time explicitly incorporates Rotary Position Embedding (RoPE) to construct a structured spatial field. The method innovatively extends RoPE into a depth-aware formulation to encode 3D spatial relationships. Combined with synthetic data pretraining and parameter-efficient fine-tuning, the model achieves strong object identity preservation and coherent scene updates under large displacements, even with minimal real-data supervision. It significantly outperforms prior art on standard benchmarks, generating plausible novel content while consistently adapting illumination and shadow effects.

Abstract

Moving an object in a single image requires geometry-consistent spatial rearrangement, including handling occlusions, revealing previously unseen regions, and maintaining coherent shadows and reflections. Existing approaches are not well suited to this setting and often fail to preserve such scene-level consistency. We address this problem by introducing a geometry-aware object motion method that operates directly on the positional representations of diffusion transformers. Our key insight is that rotary positional embeddings (RoPE) define a structured spatial field that can be explicitly manipulated to induce controlled motion. We extend 2D RoPE into a depth-aware formulation that encodes 3D spatial structure, enabling consistent object displacement and scene-aware updates. Our model is trained using synthetic data combined with a small set of real images via parameter-efficient fine-tuning. Despite minimal real supervision, it preserves object identity under large spatial displacements, generates plausible content in newly revealed regions, and consistently updates scene-dependent effects such as shadows and illumination. Experimental results on standard object motion benchmarks demonstrate state-of-the-art performance across all evaluation metrics.

Problem

• object relocation
• geometry consistency
• occlusion handling
• scene-level coherence
• image manipulation

Innovation

• RoPE
• depth-aware
• object relocation
• diffusion transformer
• positional embeddings