ECCV 2026
Zesong Yang*, Yuanhang Lei*, Liyuan Cui, Yihang Chen, Jiaer Huang,
Boming Zhao, Peter Yichen Chen, Hujun Bao, Zhaopeng Cui†
- [2026-10-30] MeGAS is now featured in PhysStudio, our open-source world simulator for physics-based animation. 🌏
- [2026-10-08] The MeGAS simulation code and interactive viewer are now available! Welcome to try them out and share your feedback.
- [2026-06-17] MeGAS has been accepted to ECCV 2026! 🎉
Recent advances integrate physically grounded Newtonian dynamics with neural rendering frameworks, narrowing the gap between photorealistic scene reconstruction and physics-based animation. However, existing approaches focus on mechanically driven dynamics while neglecting temperature, a fundamental yet invisible physical factor underlying phenomena such as melting, solidification, and other thermomechanical processes.
In this paper, we propose MeGAS, a novel framework that incorporates thermomechanical phase-change dynamics into 3D Gaussian Splatting (3DGS). Specifically, we propose a new thermomechanical dynamic Gaussian Splatting representation that augments 3DGS with temperature attributes and employs a heat advection-diffusion solver with MPM dynamics incorporating phase transitions, enabling physically plausible and visually realistic synthesis of thermophysical phenomena.
Furthermore, a new topology-adaptive Gaussian rendering strategy is proposed to mitigate cracking and floaters under extreme deformation. Extensive experiments demonstrate that MeGAS produces physically consistent thermomechanical behavior while maintaining high-fidelity photorealistic rendering, advancing toward physics-integrated world models.
Tested with PyTorch 2.0 · CUDA 12.4 · NVIDIA Warp 1.7.2 on an NVIDIA RTX 4070 Ti (12 GB). After installing PyTorch:
bash run_install.shSimulation. The input Gaussians are expected at log/gaussians/<scene>/align/align_gaussian.ply.
bash scripts/1_run_candy_house.shResults are written to log/gaussians/candy_house/mpm/solid-jelly_liquid-foam_thermal-sphere/test/@<timestamp>/.
Interactive viewer. Set --mpm_path to the output directory above:
python gs_simulation_gui.py --mpm_path log/gaussians/candy_house/mpm/solid-jelly_liquid-foam_thermal-sphere/test/@<timestamp>Original MeGAS code is licensed under Apache 2.0. Third-party components retain their respective licenses.
Parts of this code are adapted from PhysGaussian and Warp-MPM. We thank their authors for making these valuable resources available.
If you find this work useful for your research, please consider citing:
@inproceedings{yang2026megas,
title={MeGAS: Thermomechanical Dynamic Gaussian Splatting for Thermophysical Scene Editing},
author={Yang, Zesong and Lei, Yuanhang and Cui, Liyuan and Chen, Yihang and Huang, Jiaer and Zhao, Boming and Chen, Peter Yichen and Bao, Hujun and Cui, Zhaopeng},
booktitle={European Conference on Computer Vision},
pages={390--409},
year={2026},
organization={Springer}
}