Research map / Label-efficient and robust learning / Domain adaptation and generalization
M2EIT: Multi-Domain Mixture of Experts for Robust Neural Inertial Tracking
Yan Li, Yang Xu, Changhao Chen, Zhongchen Shi, Wei Chen, Liang Xie, Hongbo Chen, Erwei Yin
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 temporal decomposition expert (Bi-Mamba) to the spatial expert (ablation) | M2EIT · RIDI | ATE: 2.47 -> 2.05 | Improved |
| add temporal decomposition expert (Bi-Mamba) to the spatial expert (ablation) | M2EIT · RIDI | RTE: 2.56 -> 2.16 | Improved |
| add wavelet frequency decomposition expert to SDE+TDE (ablation) | M2EIT · RIDI | ATE: 2.05 -> 1.65 | Improved |
| add wavelet frequency decomposition expert to SDE+TDE (ablation) | M2EIT · RIDI | RTE: 2.16 -> 2.00 | Improved |
| use all three experts instead of SDE alone (ablation) | M2EIT · RIDI | ATE: 37.5% improvement in localization accuracy compared to SDE | Improved |
| add cross-domain translation layer (ablation) | M2EIT · RIDI | ATE and RTE | Improved |
| replace mean fusion with soft router (ablation) | M2EIT · RIDI | ATE and RTE | Improved |
| use more expert domains under input noise 0.01, 0.05, 0.1 (ablation) | M2EIT · RIDI (noise robustness) | ATE and RTE | Improved |
| use M2EIT instead of RoNIN ResNet | M2EIT · RIDI (seen) | ATE: 2.33 -> 1.65 | Improved |
| use M2EIT instead of RoNIN ResNet | M2EIT · RoNIN (seen) | RTE: 2.87 -> 2.76 | Improved |
| use M2EIT instead of RoNIN ResNet | M2EIT · RoNIN (unseen) | RTE: 4.65 -> 4.57 | No clear effect |
Closest papers by meaning
- MotionTransformer: Transferring Neural Inertial Tracking between Domains (AAAI 2019 · similarity 0.87)
- EqNIO: Subequivariant Neural Inertial Odometry (ICLR 2025 · similarity 0.84)
- MODA: Motion-Drift Augmentation for Inertial Human Motion Analysis (CVPR 2025 · similarity 0.84)
- Wavelet Dynamic Selection Network for Inertial Sensor Signal Enhancement (AAAI 2024 · similarity 0.84)
- iMoT: Inertial Motion Transformer for Inertial Navigation (AAAI 2025 · similarity 0.82)
- Spatial-Related Sensors Matters: 3D Human Motion Reconstruction Assisted with Textual Semantics (AAAI 2024 · similarity 0.82)
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.
