Using Microwave Simulations for the Generation of Radar Data for Gesture and Activity Recognition: Potential and Challenges for Real-World Applications
- 7 October 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Microwave Magazine
- Vol. 25 (11) , 18-29
Abstract
In recent years, algorithms have been successfully developed to recognize a wide range of activities and gestures with millimeter-wave radar sensors. Applications range from patient monitoring in health care [1] to contactless gesture control of smart devices [2] and traffic gesture recognition by autonomous vehicles [3] . All these approaches involve machine learning (ML) models that first extract valuable information from the radar data and then use it to make a prediction about the observed human motion. The design of the involved neural networks often requires the acquisition of an application-specific training dataset. Since the acquisition of large amounts of gesture or activity observations is very laborious, typical radar datasets have a few hundred to a few thousand samples per class [4] , [5] , [6] , [7] . The small number of samples per class prevents the use of very deep neural networks with many layers and millions of trainable parameters. If the neural network becomes too powerful for a given dataset size, it will start to simply memorize the data from training and be unable to correctly classify new samples, a problem known as overfitting . The authors of [8] and [9] train a convolutional neural network (CNN) with increasing depth on a small activity dataset and observe severe overfitting as the model becomes too powerful. This prevents the further increase in accuracy that would be expected from deeper neural networks.Keywords
This publication has 51 references indexed in Scilit:
- Synthetic Radar Dataset Generator for Macro-Gesture RecognitionIEEE Access, 2021
- High Fidelity Physics Simulation of 128 Channel MIMO Sensor for 77GHz Automotive RadarIEEE Access, 2020
- VIBE: Video Inference for Human Body Pose and Shape Estimation2020
- Machine Learning-Based Radar Perception for Autonomous Vehicles Using Full Physics SimulationIEEE Access, 2020
- Short-Range Radar-Based Gesture Recognition System Using 3D CNN With Triplet LossIEEE Access, 2019
- Training Deep Networks with Synthetic Data: Bridging the Reality Gap by Domain Randomization2018
- Learning from Synthetic Data: Addressing Domain Shift for Semantic Segmentation2018
- Hand Keypoint Detection in Single Images Using Multiview Bootstrapping2017
- Microdoppler signature simulation of computer animated human and animal motions2008
- A global human walking model with real-time kinematic personificationThe Visual Computer, 1990