Generating Synthetic Augmentation Data from a Practical UWB Radar Dataset Using VQ-VAE
Bibliographic record
Abstract
Human Activity Recognition (HAR) is a field that has attracted significant attention due to its wide range of applications, from healthcare monitoring to smart home systems. Because of the aging of the population in first-world countries, it becomes increasingly important to find innovative solutions that reduce risks associated with aging-in-place policies. However, development of robust HAR systems is faced with two main challenges: the lack of generalization capacities of current methods and the lack of sufficient practical labeled data. Traditional data augmentation techniques do improve classification accuracy, but they often don’t fully solve the complexity and variability of human activities. This article introduces an innovative approach to creating synthetic data for HAR using Vector-Quantized Variational Auto Encoders (VQ-VAEs), aiming to address the challenges of data scarcity and improve the ability of HAR systems to generalise. Our approach includes the use of three UWB radars to recognize 14 activities performed by 19 participants in a prototype smart-home apartment as the practical dataset. Data from those three UWB radars are then filtered using a Variational Auto Encoder (VAE) and used as ground for data augmentation. Following best practices in data separation, the Leave-One-Subject-Out (LOSO) evaluation method validates the data augmentation using the MobileNetV2 classifier architecture. Results show that the quality of generated data is similar to the original dataset, potentially enhancing the performance of UWB radar based HAR systems.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".