English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

SimuScene: Simulation-Ready Compositional 3D Scene Reconstruction from a Single Image

Forum topic · 小凯 · 2026-06-04

Summary

SimuScene (arXiv:2606.03994) is a computer vision paper by Inhee Lee, Sangwon Baik, Sungjoo Kim, Hyeonwoo Kim, Hyunsoo Cha, and Hanbyul Joo that tackles reconstructing interactive, simulation-ready 3D scenes from a single image — a key bottleneck for robotic manipulation. While recent single-image lifters recover plausible per-object shapes, composed scenes often collapse under physical simulation due to interpenetrating, hovering, or sinking objects. Existing physics-aware methods treat this only as post-hoc layout correction, leaving underlying geometric errors unresolved. SimuScene instead integrates physics into the loop of shape and layout estimation, using the physics engine as a diagnostic measurement tool during the generative process. Diagnostic simulation under gravity converts penetration and support failures into quantitative correction signals that drive gravity-axis stretching and amodal shape resampling, mitigating accumulated reconstruction errors. Experiments show state-of-the-art performance on physical stability and geometric alignment benchmarks.

Overview

Research area: Computer Vision Authors: Inhee Lee, Sangwon Baik, Sungjoo Kim, Hyeonwoo Kim, Hyunsoo Cha, Hanbyul Joo Published: 2026-06-02 arXiv: 2606.03994

Summary

Reconstructing interactive, simulation-ready 3D scenes from a single image is a critical bottleneck for robotic manipulation. While recent single-image lifters recover plausible per-object shapes, composing them yields scenes that collapse under physical simulation due to interpenetrating, hovering, or sinking objects. Existing physics-aware methods address this strictly as a post-hoc layout correction, leaving the underlying geometric errors unresolved.

To address this, the authors introduce SimuScene, a compositional 3D reconstruction pipeline that puts physics in the loop of shape and layout estimation. Rather than using physics merely for layout cleanup, the system utilizes the physics engine as a diagnostic measurement tool during the generative process itself. By diagnostically simulating reconstructed objects under gravity, SimuScene converts penetration and support failures into quantitative correction signals that drive gravity-axis stretching and amodal shape resampling.

This physics-informed feedback loop mitigates accumulated reconstruction errors, producing stable, simulation-ready compositional 3D scenes. Extensive experiments demonstrate state-of-the-art performance on physical stability and geometric alignment benchmarks.

Key contributions

  • Physics integrated into the shape and layout estimation loop, not just post-hoc layout cleanup
  • Physics engine used as a diagnostic measurement tool during generation
  • Penetration and support failures converted into quantitative correction signals (gravity-axis stretching, amodal shape resampling)
  • State-of-the-art results on physical stability and geometric alignment benchmarks

Tags

#computer-vision#3d-reconstruction#physics-simulation#robotics#scene-understanding#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177980804