Thermal Fall 66: A robust dataset for thermal imaging-based fall detection and eldercare
Bibliographic record
Abstract
Fall-related injuries pose a significant health risk, particularly among the elderly population, necessitating advancements in fall detection technologies. While traditional sensor-based methods offer some solutions, they are limited by issues, like user compliance and unreliability in high-stakes situations. Computer vision (CV) and artificial intelligence (AI)-based alternatives, notably thermal imaging, emerge as promising, privacy-preserving tools. However, the lack of standardized, generalizable benchmark datasets hampers the validation of these technologies’ effectiveness. The Thermal Fall 66 (TF-66) dataset introduced in this work addresses this crucial gap by providing an extensive, diverse collection of fall scenarios, recorded in various environments and featuring a wide range of participants. This dataset introduces sample subsets for tailored model training and a comprehensive data generator for customized access, setting a new benchmark in fall detection research. A baseline three-dimensional convolutional neural network (3D CNN) was trained to demonstrate the dataset’s suitability for supervised deep learning. As TF-66 continues to evolve, it aims to include even more diverse environments, such as bathrooms and staircases, further enhancing its applicability and serving as a robust resource for the research community. This work contributes significantly to the field by offering a dataset that not only improves the reliability of fall detection systems (FDS) but also facilitates a standardized approach to evaluating and advancing these technologies.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 teacher head, 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".