Data-efficient multimodal human action recognition for proactive human-robot collaborative assembly: A cross-domain few-shot learning approach.
Tianyu WangZhihao LiuLihui WangMian LiXi Vincent WangPublished in: Robotics Comput. Integr. Manuf. (2024)
Keyphrases
- action recognition
- cross domain
- computer vision
- human robot
- domain adaptation
- prior knowledge
- knowledge transfer
- learning algorithm
- action classification
- human subjects
- human activities
- data points
- training data
- data mining
- multi modal
- collaborative filtering
- human actions
- supervised learning
- learning process
- video sequences