AI-Driven Slip Detection for Smarter Footwear Testing using Vision Transformers
Bibliographic record
Abstract
Slips and falls remain significant global health concerns, contributing to injuries, loss of independence, and increasing healthcare costs. Among the primary causes, slipping on icy outdoor surfaces during winter poses a significant risk due to hazardous conditions. The use of slip-resistant footwear has been shown to significantly reduce the incidence of slips and falls. Therefore, in this paper, we aim to enhance footwear slip-resistance testing protocols to ensure more accurate, reliable, and consistent evaluation of footwear slip-resistant performance. We introduced a novel automated slip detection system to overcome the inconsistencies and errors often linked to human observation during human-centered footwear testing. By incorporating AI-based methods, our system eliminated subjective bias and improved the consistency and precision of slip resistance assessments, particularly for winter footwear. This study conducted transfer learning using two pre-trained neural network architectures: Inflated 3D ConvNets (I3D) and Multiscale Vision Transformers Version 2 (MViTv2). The models were trained on 410 walking trials conducted on a controlled icy walkway with adjustable slope angles, ensuring a balanced dataset of both slip and non slip events. The slip detection models were evaluated using 5-Fold and Leave-One-Subject-Out (LOSO) cross-validation techniques. Although the performance difference between the two models was not statistically significant, MViTv2 outperformed I3D with an average accuracy of 90.49% ± 2.93% and 88.80% ± 4.82% for 5-Fold and LOSO, respectively. This model also demonstrated greater consistency in performance compared to I3D. Our findings highlighted the potential of video-based AI models, particularly vision transformers, for precise automated slip detection. The study demonstrated the feasibility of integrating these models into footwear testing protocols, paving the way for improved safety in icy environments.Clinical relevance- Implementing AI-based slip detection in footwear testing protocols can contribute to improved fall prevention strategies, ultimately reducing healthcare costs and enhancing mobility and independence in icy environments.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".