Geometry-Editable and Appearance-Preserving Object Compositon

TL;DR

The paper introduces DGAD, a diffusion-based model that enables geometry-editable and appearance-preserving object composition by combining semantic embeddings with a cross-attention retrieval mechanism.

Contribution

DGAD is the first model to explicitly disentangle geometry editing and appearance preservation in object composition using diffusion models and cross-attention.

Findings

DGAD achieves superior geometric editing and appearance preservation on benchmarks.

The cross-attention retrieval mechanism effectively aligns fine-grained appearance features.

Experiments demonstrate DGAD's ability to produce realistic, geometry-adjusted composite images.

Abstract

General object composition (GOC) aims to seamlessly integrate a target object into a background scene with desired geometric properties, while simultaneously preserving its fine-grained appearance details. Recent approaches derive semantic embeddings and integrate them into advanced diffusion models to enable geometry-editable generation. However, these highly compact embeddings encode only high-level semantic cues and inevitably discard fine-grained appearance details. We introduce a Disentangled Geometry-editable and Appearance-preserving Diffusion (DGAD) model that first leverages semantic embeddings to implicitly capture the desired geometric transformations and then employs a cross-attention retrieval mechanism to align fine-grained appearance features with the geometry-edited representation, facilitating both precise geometry editing and faithful appearance preservation in object composition. Specifically, DGAD builds on CLIP/DINO-derived and reference networks to extract semantic embeddings and appearance-preserving representations, which are then seamlessly integrated into the encoding and decoding pipelines in a disentangled manner. We first integrate the semantic embeddings into pre-trained diffusion models that exhibit strong spatial reasoning capabilities to implicitly capture object geometry, thereby facilitating flexible object manipulation and ensuring effective editability. Then, we design a dense cross-attention mechanism that leverages the implicitly learned object geometry to retrieve and spatially align appearance features with their corresponding regions, ensuring faithful appearance consistency. Extensive experiments on public benchmarks demonstrate the effectiveness of the proposed DGAD framework.