Training Lightweight CNNs for Human-Nanodrone Proximity Interaction from Small Datasets using Background Randomization
2021
We consider the task of visually estimating the pose of a human from images acquired by a nearby nano-drone; in this context, we propose a data augmentation approach based on synthetic background substitution to learn a lightweight CNN model from a small real-world training set. Experimental results on data from two different labs proves that the approach improves generalization to unseen environments.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
22
References
0
Citations
NaN
KQI