Tracking-based 3D human skeleton extraction from stereo video camera toward an on-site safety and ergonomic analysis
Bibliographic record
Abstract
Purpose As a means of data acquisition for the situation awareness, computer vision-based motion capture technologies have increased the potential to observe and assess manual activities for the prevention of accidents and injuries in construction. This study thus aims to present a computationally efficient and robust method of human motion data capture for the on-site motion sensing and analysis. Design/methodology/approach This study investigated a tracking approach to three-dimensional (3D) human skeleton extraction from stereo video streams. Instead of detecting body joints on each image, the proposed method tracks locations of the body joints over all the successive frames by learning from the initialized body posture. The corresponding body joints to the ones tracked are then identified and matched on the image sequences from the other lens and reconstructed in a 3D space through triangulation to build 3D skeleton models. For validation, a lab test is conducted to evaluate the accuracy and working ranges of the proposed method, respectively. Findings Results of the test reveal that the tracking approach produces accurate outcomes at a distance, with nearly real-time computational processing, and can be potentially used for site data collection. Thus, the proposed approach has a potential for various field analyses for construction workers’ safety and ergonomics. Originality/value Recently, motion capture technologies have rapidly been developed and studied in construction. However, existing sensing technologies are not yet readily applicable to construction environments. This study explores two smartphones as stereo cameras as a potentially suitable means of data collection in construction for the less operational constrains (e.g. no on-body sensor required, less sensitivity to sunlight, and flexible ranges of operations).
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| 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) | 0.002 | 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".