Science & Tech · AI · 1 hr ago
Noitom Robotics Releases Hand-Movement Dataset Used for NVIDIA Model
Noitom Robotics released the HiPHI-Hand dataset on Hugging Face.
It contains 11 hours of stereo video filmed from the wearer’s perspective, showing hand movements.
The dataset includes marker-based ground truth to identify where hands are in the footage.
NVIDIA used HiPHI-Hand as one source for its EgoHand model.
This connects the dataset to work on models that interpret hands in first-person video.
The information provided does not say what happens next.
Noitom Robotics released the HiPHI-Hand dataset on Hugging Face.
The dataset contains 11 hours of egocentric stereo video.
It includes marker-based ground truth for hands.
NVIDIA used the dataset as one source to train its EgoHand model.
- Who
- Noitom Robotics released the dataset; NVIDIA used it to train EgoHand.
- What
- Noitom Robotics released HiPHI-Hand, a hand-movement dataset.
- When
- November 10, 2026.
- Where
- On Hugging Face.
- Why
- The dataset was used as one source to train NVIDIA’s EgoHand model.
This story does not have two clearly opposing sides.
This story does not have a timeline yet.
- Dataset
- HiPHI-Hand
- Release platform
- Hugging Face
- Video duration
- 11 hours
- Video type
- Egocentric stereo video
- Ground truth
- Marker-based hand ground truth
- Model
- NVIDIA’s EgoHand model