A model-based fusion technique for ocular motion sensing
Bibliographic record
Abstract
Electrooculography (EOG) uses an electrical signal that records the cornea-retinal differential potential of the eyeball, which is linearly proportional to the eye movement. In EOG, the electrodes placed in the outer canthus and lateral frontalis record the horizontal and vertical eye movements. A main challenge in these signals is that the eye movement information is compressed in the 0-35Hz range of the full spectrum of the recording (0-250/1000/2000Hz) depending on the dynamic range of the device. Moreover, these signals, like any other bio-signals, are contaminated heavily by artifacts and noises such as electroencephalography (EEG), eye-lid motions, illumination drift, skin drift, impedance drop of electrode, eye blinks, head movements, and DC drift. Researchers have attempted to addressing these issues using traditional techniques, including finite impulse response (FIR) filters, morphological filters, time series motifs and wavelet-based approaches. However, most of these approaches have not been able to improve the performance significantly without compromising other features of the system such as the computational cost, real-time performance and time lag. This thesis presents a model-based fusion technique for ocular motion sensing to improve the signal quality. The developed approach is tested with five different model-based approaches (Brownian, constant velocity, constant acceleration, Westheimer, and linear reciprocal). Among these, the approach based on the linear reciprocal human eye model show a significant overall improvement in the signal(500-700% improvement in SNR w.r.t. FIR and 40-50% decrement in computational cost). Therefore, it is concluded that the approach based on the linear reciprocal model provides the best fit in the implementation of high quality eye tracking systems.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| 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".