Training Lightweight CNNs for Human-Nanodrone Proximity Interaction from Small Datasets using Background Randomization
M Ferri, D Mantegazza, E Cereda… - arXiv preprint arXiv …, 2021 - arxiv.org
arXiv preprint arXiv:2110.14491, 2021•arxiv.org
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.
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.
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.
arxiv.org
以上显示的是最相近的搜索结果。 查看全部搜索结果