PhysMirror: Physics-Aware Mirror Object Generation

2026-07-03

AI Summary

This work addresses the challenge that current text-to-image diffusion models struggle to satisfy strict geometric constraints when generating mirror reflections, often producing physically implausible or hallucinated results. To overcome this limitation, the authors propose the first end-to-end physics-aware generation framework that leverages explicit 3D spatial priors to automatically construct mirror scenes and uses rendered depth and segmentation maps to guide the diffusion model toward generating physically consistent mirror images. The core innovation lies in integrating explicit 3D geometric modeling with diffusion-based synthesis and introducing Mirror Consistency Score (MCS), a reference-free automated metric for evaluating reflection fidelity. Experiments on the newly introduced MirrOB dataset demonstrate that the proposed method significantly outperforms existing approaches in terms of reflection accuracy, physical realism, and alignment with input text semantics.

Abstract

Synthesizing physically accurate mirror reflections remains a fundamental challenge for modern text-to-image diffusion models, which are increasingly critical for generating synthetic training data for embodied AI and robotic perception. These models typically struggle with strict geometric constraints, leading to hallucinations that degrade the utility of the synthetic data. To address this, we introduce a novel, end-to-end physics-aware generation framework namely PhysMirror that natively enforces projective geometry through explicit 3D spatial priors. Our method automatically lifts prompted objects into 3D meshes and constructs a lightweight, mathematically exact mirror scene within a simulated environment. By rendering this explicit 3D scene, we extract precise 2D conditioning elements, such as depth maps and segmentation maps, that serve as robust guiding signals for downstream diffusion models, guiding them to generate images with physically correct mirror reflections. Moreover, we introduce Mirror Consistency Score (MCS), reference-free, fully automated metric that quantifies physical correctness using dense feature matching and vanishing point convergence. Experimental results on our newly constructed MirrOB dataset demonstrate that our approach outperforms state-of-the-art baselines in reflection accuracy and physical realism, while maintaining strong text-to-image semantic alignment, providing a reliable pipeline for embodied AI data generation. The source code is released at https://duyphuc0701.github.io/PhysMirror.

Problem

• mirror reflection
• physical accuracy
• geometric constraints
• synthetic data generation
• embodied AI

Innovation

• physics-aware generation
• mirror reflection synthesis
• 3D spatial priors
• diffusion models
• Mirror Consistency Score