Training Lightweight CNNs for Human-Nanodrone Proximity Interaction from Small Datasets using Background Randomization

10/27/2021

∙

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.

READ FULL TEXT

Training Lightweight CNNs for Human-Nanodrone Proximity Interaction from Small Datasets using Background Randomization

Towards Visible and Thermal Drone Monitoring with Convolutional Neural Networks

PromptMix: Text-to-image diffusion models enhance the performance of lightweight networks

A Lightweight Privacy-Preserving Scheme Using Label-based Pixel Block Mixing for Image Classification in Deep Learning

Selective Synthetic Augmentation with Quality Assurance

Spatio-temporal Data Augmentation for Visual Surveillance

Domain Adaptive 3D Pose Augmentation for In-the-wild Human Mesh Recovery

CNN-based local features for navigation near an asteroid

Training Lightweight CNNs for Human-Nanodrone Proximity Interaction from Small Datasets using Background Randomization

Related Research

Towards Visible and Thermal Drone Monitoring with Convolutional Neural Networks

PromptMix: Text-to-image diffusion models enhance the performance of lightweight networks

A Lightweight Privacy-Preserving Scheme Using Label-based Pixel Block Mixing for Image Classification in Deep Learning

Selective Synthetic Augmentation with Quality Assurance

Spatio-temporal Data Augmentation for Visual Surveillance

Domain Adaptive 3D Pose Augmentation for In-the-wild Human Mesh Recovery

CNN-based local features for navigation near an asteroid