Research map / 3D vision / Gaussian splatting
HybridSim: A Physics–Learning Hybrid Digital Twin for mmWave Human Sensing
Weitao Xiong, Tianyu Liu, Peng Li, KOK CHUNG CHUA, Chean Khim Toa, Pu Wang, Hongfei Xue
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add decoupled BRDF-based direct path instead of coupled 3DGS (ablation) | HybridSim · unseen subject | PSNR (dB): 19.12 -> 22.29 | Improved |
| add decoupled BRDF-based direct path instead of coupled 3DGS (ablation) | HybridSim · HAR | accuracy (%): 85.61 -> 92.07 | Improved |
| add 3DGS indirect path (ablation) | HybridSim · HAR | accuracy (%): 88.96 -> 92.07 | Improved |
| use 3DGS indirect path instead of two-bounce ray tracing (ablation) | HybridSim · HAR | accuracy (%): 41.6 -> 92.07 | Improved |
| use 3DGS indirect path instead of two-bounce ray tracing (ablation) | HybridSim · per-frame runtime | seconds per frame: 3.44s -> under 1s | Improved |
| add fixed noise floor (ablation) | HybridSim · HAR | accuracy (%): 72.79 -> 90.51 | Improved |
| use learnable noise instead of fixed noise (ablation) | HybridSim · HAR | accuracy (%): 90.51 -> 92.07 | Improved |
| disable GCN at inference (ablation) | HybridSim · HAR | accuracy (%): 86.79 -> 92.07 | Improved |
| disable GCN at inference (ablation) | HybridSim · RD synthesis (global) | PSNR (dB): 22.292 -> 22.294 | No clear effect |
| disable GCN at inference (ablation) | HybridSim · Action 6 static-removed | PSNR (dB): 22.08 -> 22.41 | Improved |
| add GCN with stochastic masking during training (ablation) | HybridSim · training stability | loss spikes | Improved |
| use HybridSim instead of mmGPE | simulator · HAR | accuracy (%): 54.22 -> 92.07 | Improved |
Closest papers by meaning
- M4Human: A Large-Scale Multimodal mmWave Radar Benchmark for Human Mesh Reconstruction (CVPR 2026 · similarity 0.86)
- MVDoppler-Pose: Multi-Modal Multi-View mmWave Sensing for Long-Distance Self-Occluded Human Walking Pose Estimation (CVPR 2025 · similarity 0.82)
- MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception (NeurIPS 2026 · similarity 0.82)
- 3D Point Splatting for mmWave Radar Novel View Synthesis (NeurIPS 2026 · similarity 0.81)
- NEAR: Neural Electromagnetic Array Response (ICML 2025 · similarity 0.81)
- Physics Unrolled Neural Operator for Wireless Field Modeling (NeurIPS 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
