Automation in Construction· 2026Q1
Synthetic data-driven framework for estimating worker activity intensity using pose-based graph neural networks
- 0citations
- Q1SCImago
- 2026year
Short summary
A framework using synthetic images and graph neural networks (GNNs) accurately estimates construction workers' metabolic equivalent of task (MET) classes, with a 5-frame window improving predictions over single frames.
AI-generated from the title and abstract; the full text is not read.
Key points
- Developed a framework combining synthetic image generation (Stable Diffusion) and skeleton-based GNNs (B-ResSkelGCN) for MET class prediction.
- Synthetic images were generated using domain-informed prompts and processed into GCN inputs.
- A 5-frame window integration improved MET class prediction accuracy and reduced error compared to single-frame predictions.
- Prediction stability varied with window length, indicating a need for adaptive integration strategies.
AI-generated from the title and abstract; the full text is not read.
Abstract
This paper proposes a metabolic equivalent of task (MET) class prediction framework that combines Stable Diffusion-generated synthetic images with B-ResSkelGCN, which learns from skeleton graphs, for non-contact evaluation of construction workers' workload levels. Using domain-informed prompts, synthetic images were generated, followed by worker region of interest (ROI) alignment and skeleton transformation to construct graph convolutional network (GCN) inputs. The trained B-ResSkelGCN achieved high MET class prediction performance on frames in a test environment. However, predictions from single frames in real videos fluctuated during action transitions or momentary stillness, such as walking. To address this issue, mode-based window integration aggregated predictions within each window. A 5-frame window improved classification performance and MET error metrics compared with single-frame prediction, whereas excessively long windows diluted action-transition information and increased confusion, highlighting the need for operational strategies that adjust integration intervals according to action-transition characteristics and management objectives.
The authors' abstract, as published at the source. Automation in Construction, 2026 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Radiological and Ultrasound Technology
Radiological and Ultrasound TechnologyHealth Professions