TouchScale: 500 Hours of Human Vision and Touch for Visual-Tactile Learning
Large-scale egocentric human interaction data is becoming an important source of physical supervision for embodied learning, yet video alone leaves the contact and pressure that characterize physical interaction unrecorded.
ProofPaper ↗
Key points
- We therefore introduce TouchScale, a 500-hour dataset of contact-rich human interaction recorded with a single unified wearable setup.
- Compared with prior tactile data, training on the full TouchScale raises zero-shot contact IoU on data from an unseen tactile sensor from 0.134 to 0.383.
- These results suggest that human visual-tactile data collected at scale with consistent sensing benefits both perception and robot manipulation.
- We will publicly release TouchScale, including all synchronized visual-tactile recordings and reconstructed object models, to support future research on scalable visual-tactile learning.
Sources (1)
- [1]TouchScale: 500 Hours of Human Vision and Touch for Visual-Tactile LearningarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 03:49 PM
Large-scale egocentric human interaction data is becoming an important source of physical supervision for embodied learning, yet video alone leaves the contact and pressure that characterize physical interaction unrecorded.
We therefore introduce TouchScale, a 500-hour dataset of contact-rich human interaction recorded with a single unified wearable setup.
Extractive summary: sentences quoted from the sources.