Back to Search
Start Over
Training Lightweight CNNs for Human-Nanodrone Proximity Interaction from Small Datasets using Background Randomization
- Publication Year :
- 2021
-
Abstract
- 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.
Details
- Database :
- arXiv
- Publication Type :
- Report
- Accession number :
- edsarx.2110.14491
- Document Type :
- Working Paper