Downloads · 30 days
6
43% of all-time downloads
imu_odometry_challenge_iros26
haozheee/imu_odometry_challenge_iros26
imu_odometry_challenge_iros26 is a machine learning model from haozheee. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Single unified model for IMU-only body-frame velocity estimation across four platforms (car / quadruped / drone / human).
All-time downloads
14
Public
Repo size
51.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt51.6 MB · 97%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Sep 20, 2026
- Updated
- Sep 20, 2026
- SHA
- f8405c04
Model card
From the Hugging Face model README
Submission to TartanIMU Challenge (IROS 2026)
Single unified model for IMU-only body-frame velocity estimation across four platforms (car / quadruped / drone / human).
Contents
| File | Description |
|---|---|
model.pt | Frozen weights of the single unified model |
config.json | Architecture settings of model.pt |
model.py | Model definition |
infer.py | Self-contained inference entry point (no internet, no external repo imports) |
submission.csv | The exact prediction file that produced the scores above (md5 b6d8a00027780279b046534ea8ca20f1) |
requirements.txt | Dependencies (torch / numpy / pandas only); requirements-lock.txt has the exact versions used |
How to run
pip install -r requirements.txt
python infer.py --test_dir /path/to/challenge_data/test \
--windows /path/to/challenge_data/index/test_windows.csv \
--out submission.csv
Method
We use an encoder–decoder transformer that regresses body-frame velocity from raw IMU signals. The input is processed by a strided convolutional stem plus a multi-resolution log-spectrogram branch to derive the IMU tokens. The context is compressed into an embodiment state vector that adapts the LayerNorm parameters to the embodiment. For data augmentation, we re-time and spatially stretch the training trajectories while keeping the IMU signals physically consistent. Details are in the technical report.
Result
Official scoring service, all 89 test sequences (submission md5 b6d8a00027780279b046534ea8ca20f1).
| Platform | ATE20 (m) | AVE (m/s) | RTE@5s (m) |
|---|---|---|---|
| Wheeled / Car | 0.38504 | 0.06799 | 0.30095 |
| Handheld / Human | 0.53125 | 0.05126 | 0.29518 |
| Legged / Quadruped | 0.39942 | 0.07339 | 0.36021 |
| Aerial / Drone | 0.93501 | 0.50103 | 2.31594 |
| Average | 0.56268 | 0.17342 | 0.81807 |
TartanIMU Score 0.21367 · Kaggle public leaderboard 0.28283
Compliance Declaration
- Do the predictions come from a single model with one shared set of weights? Yes. One model with one set of shared weights.
- Is any platform classification, routing or specialization used at inference? No. There is no platform label, classifier, routing or per-platform expert at inference. The LayerNorms are adapted automatically at inference time as an internal behavior of the model. However, at training time, the platform label is used as an auxiliary supervision signal.
- Is any ensembling, checkpoint averaging or test-time augmentation used? No.
- How was test-set leakage avoided, including for model and checkpoint selection?
The test IMU was never used for training or adaptation. Design choices were made on the released
valsplit. We made very few submissions against the test set, so the risk of overfitting to the test data is low. - What differs between your development-time system and the submitted system?
For development, we performed architecture search on the original split of
trainandval. The submitted model is trained ontrain+valcombined, with a longer schedule. The training-only auxiliary heads are removed from the released weights.
Contact
Haozhe Zhou, [email protected], Carnegie Mellon University
- License
- apache-2.0
- Created
- Sep 20, 2026
- Updated
- Sep 20, 2026