Dataset · IntelLabs
Multimodal robotic welding data
IntelLabs/Intel_Robotic_Welding_Multimodal_Dataset
Over 4k annotated welding samples: video, audio, sensor time series, and five seam images.
What's inside
Collected to study welding defect detection in a multimodal setup. Contains over 4,000 annotated samples from an automotive plant production line gathered with a supplier. Each sample includes video, audio, welding sensor time series, and five post-welding seam images.
What you can do with it
- Training a multimodal welding defect detection model
- Testing video, audio, and time-series fusion
- Evaluating model transfer to industrial data
- Downloads this month
- 423
- Likes
- 99
- Access
- on request on Hugging Face
Models trained on it
How to load it
from datasets import load_dataset
ds = load_dataset("IntelLabs/Intel_Robotic_Welding_Multimodal_Dataset")