Development and evaluation of a hand tracker using depth images captured from an overhead perspective
Bibliographic record
Abstract
PURPOSE: We present the development and evaluation of a robust hand tracker based on single overhead depth images for use in the COACH, an assistive technology for people with dementia. The new hand tracker was designed to overcome limitations experienced by the COACH in previous clinical trials. METHODS: We train a random decision forest classifier using ∼5000 manually labeled, unbalanced, training images. Hand positions from the classifier are translated into task actions based on proximity to environmental objects. Tracker performance is evaluated using a large set of ∼24 000 manually labeled images captured from 41 participants in a fully-functional washroom, and compared to the system's previous colour-based hand tracker. RESULTS: Precision and recall were 0.994 and 0.938 for the depth tracker compared to 0.981 and 0.822 for the colour tracker with the current data, and 0.989 and 0.466 in the previous study. CONCLUSIONS: The improved tracking performance supports integration of the depth-based tracker into the COACH toward unsupervised, real-world trials. Implications for Rehabilitation The COACH is an intelligent assistive technology that can enable people with cognitive disabilities to stay at home longer, supporting the concept of aging-in-place. Automated prompting systems, a type of intelligent assistive technology, can help to support the independent completion of activities of daily living, increasing the independence of people with cognitive disabilities while reducing the burden of care experienced by caregivers. Robust motion tracking using depth imaging supports the development of intelligent assistive technologies like the COACH. Robust motion tracking also has application to other forms of assistive technologies including gaming, human-computer interaction and automated assessments.
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 | 0.001 | 0.002 |
| 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 | 0.002 |
| 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".