Public, reproducible inference for the Tartan IMU Challenge (IROS 2026). Task. Given 6-axis IMU windows (acc 3 + gyro 3, length 200) covering a motion window plus its temporal context, predict body-frame velocity (vx, vy, vz) for all four platforms (car / dog…
Model Card
Public, reproducible inference for the Tartan IMU Challenge (IROS 2026). Task. Given 6-axis IMU windows (acc 3 + gyro 3, length 200) covering a motion window plus its temporal context, predict body-frame velocity (vx, vy, vz) for all four platforms (car / dog / drone / human) with a single unified model. Model. lihuaimuiros2026v1: a shared ResNet1D encoder with an FFT branch, followed by a Transformer over a triplet of consecutive windows (prev / curr / next), and a 4-expert Mixture-of-Experts head. The expert gate is a soft router driven purely by IMU-derived features — experts are learned motion regimes (general / slow / fast / agile), not platforms. A single frozen checkpoint (seed 0) is…
Excerpt from the card by Li Hua.
Identity and Version
- Repository
- Cr0wdfade/lihua_imu_iros2026_v1
- Publisher
- Li Hua
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 0fb1e3b46bc238134c646734c70a0bd8f50d70ba
- First published
- 2026-09-18
- Last updated
- 2026-09-18
Files and Weights
8 files, 14.3 MB in total. The weights are 1 file totalling 12.9 MB in pt.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| weights/seed0.pt | Weights | 12.9 MB | 115a69552835 |
| inference.py | Configuration | 6.9 KB | — |
| model.py | Configuration | 6.5 KB | — |
| weights/manifest.json | Configuration | 86 B | — |
| README.md | Documentation | 3.0 KB | — |
| requirements.txt | Other | 263 B | — |
| submissions/submission.csv | Other | 1.3 MB | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 12.9 MB
Released by Li Hua through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 12.9 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.