Direct 3D-Aware Object Insertion via Decomposed Visual Proxies
Quick Answer
The DIRECT framework enhances object insertion by integrating interactive pose manipulation with high-fidelity 2D synthesis, outperforming prior methods in geometric controllability and visual quality.
Quick Take
The DIRECT framework enhances object insertion by integrating interactive pose manipulation with high-fidelity 2D synthesis, outperforming prior methods in geometric controllability and visual quality. It decomposes insertion into appearance, geometry, and context guidance, ensuring effective integration into target scenes. An automated data pipeline further enriches training data diversity.
Key Points
- DIRECT enables pose-controllable object insertion with high visual fidelity.
- The framework decomposes insertion into three components: appearance, geometry, and context.
- It avoids feature entanglement, preserving reference appearance while adapting to scenes.
- Automated data construction improves training data diversity and quality.
- Experiments demonstrate superior performance over previous object insertion methods.
Article Content
From source RSS / original summaryarXiv:2606. 06601v1 Announce Type: new Abstract: Object insertion aims to seamlessly composite a reference object into a specified region of a background image. Recent diffusion-based methods achieve high visual quality but formulate insertion as a simple 2D inpainting task, providing no explicit control over the object's 3D pose and limiting their practical applicability.
We propose DIRECT (Decomposed Injection for Reference Composition and Target-integration), a novel framework that integrates interactive pose manipulation with high-fidelity 2D image synthesis to enable pose-controllable object insertion. Our method decomposes the insertion conditions into three complementary components: appearance guidance capturing visual details from the reference object, geometry guidance derived from the user-adjusted 3D proxy, and context guidance from the target background.
By injecting them through separate pathways, DIRECT avoids feature entanglement and simultaneously preserves reference appearance, follows the user-specified pose, and adapts the object to the target scene. We also introduce an automated data construction pipeline to improve the diversity and quality of training data. Experiments show that DIRECT outperforms previous methods in both geometric controllability and visual quality.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.
