Metrics in a Dynamic Gaze Environment
Bibliographic record
Abstract
The development of algorithms for Areas of Interest (AOIs) segmentation represents a significant leap in dynamic gaze data analysis. These algorithms enable the transformation of two-dimensional raw data into meaningful one-dimensional metrics to unlock insights into human attention. Unlike traditional methods that focus on static gaze points in controlled environments, this research introduces novel algorithms for interpreting eye-tracking data in dynamic settings. The proposed techniques specifically address the challenge of processing gaze information in scenarios with changing frames of reference, such as driving, where understanding visual attention is critical for safety and to improve performance. This paper presents two innovative metrics designed to quantify the focus areas, duration, and scanning patterns of users in dynamic driving conditions. By employing advanced transfer learning for improved multiclass image segmentation methods, a percentage time AOI metric can be created. This metric gives insight into the amount of time a driver spends fixating on a given area. An enhancement in gaze pattern determination, achieved through class segmentation, provides a clearer depiction of drivers' true gaze patterns. This nuanced approach suggests that comprehensive attention assessment cannot rely solely on AOI metrics but requires the integration of gaze pattern derived metrics for a more accurate determination of attention-related scores. The introduction of these innovative gaze analysis methods marks a pivotal advancement in cognitive research and has practical applications. These metrics promise significant contribution to driver safety and attentiveness assessment methodologies.
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 | 0.002 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".