Research map / Label-efficient and robust learning / Domain adaptation and generalization
MotionTransformer: Transferring Neural Inertial Tracking between Domains
Changhao Chen, Yishu Miao, Chris Xiaoxuan Lu, Linhai Xie, Phil Blunsom, Andrew Markham, Niki Trigoni
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 reconstruction decoder loss (ablation) | MotionTransformer · Hand to Pocket | MSE of label prediction: 0.313 -> 0.119 | Improved |
| add reconstruction decoder loss (ablation) | MotionTransformer · Hand to Bag | MSE of label prediction: 0.335 -> 0.123 | Improved |
| add reconstruction decoder loss (ablation) | MotionTransformer · Hand to Trolley | MSE of label prediction: 0.113 -> 0.098 | Improved |
| add perceptual consistency loss (ablation) | MotionTransformer · Hand to Pocket | MSE of label prediction: 0.315 -> 0.119 | Improved |
| add perceptual consistency loss (ablation) | MotionTransformer · Hand to Bag | MSE of label prediction: 0.202 -> 0.123 | Improved |
| add perceptual consistency loss (ablation) | MotionTransformer · Hand to Trolley | MSE of label prediction: 0.140 -> 0.098 | Improved |
| domain adaptation with MotionTransformer vs source-only | MotionTransformer · Hand to Bag | MSE of label prediction: 1.129 -> 0.123 | Improved |
| MotionTransformer vs CyCADA with LSTM | MotionTransformer · Hand to Pocket | MSE of label prediction: 0.237 -> 0.119 | Improved |
| MotionTransformer vs ADDA with LSTM | MotionTransformer · Hand to Trolley | MSE of label prediction: 0.216 -> 0.098 | Improved |
| add 1K labelled target sequences (semi-supervised) (ablation) | MotionTransformer · Hand to Bag | MSE of label prediction: 0.123 -> 0.075 | Improved |
| add 1K labelled target sequences (semi-supervised) (ablation) | MotionTransformer · Hand to Pocket | MSE of label prediction: 0.119 -> 0.113 | No clear effect |
Closest papers by meaning
- iMoT: Inertial Motion Transformer for Inertial Navigation (AAAI 2025 · similarity 0.86)
- SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model (CVPR 2024 · similarity 0.85)
- Accurate and Steady Inertial Pose Estimation through Sequence Structure Learning and Modulation (NeurIPS 2024 · similarity 0.85)
- MODA: Motion-Drift Augmentation for Inertial Human Motion Analysis (CVPR 2025 · similarity 0.85)
- DiffusionPoser: Real-time Human Motion Reconstruction From Arbitrary Sparse Sensors Using Autoregressive Diffusion (CVPR 2024 · similarity 0.84)
- Universal Humanoid Motion Representations for Physics-Based Control (ICLR 2024 · similarity 0.84)
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.
