Using smooth metamers to estimate color appearance metrics for diverse <scp>color‐normal</scp> observers
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Color-normal subjects sometimes disagree about metameric matches involving highly structured SPDs, because their cone fundamentals differ slightly, but non-negligibly. This has significant implications for the design of light sources and displays, so it should be estimated. We propose a broadly applicable estimation method based on a simple adaptive "front-end" interface that can be used with any selected standard color appearance model. The interface accepts, as input, any set of color matching functions for the individual subject (for example, these could be that person's cone response functions) and also the associated tristimulus values for the test stimulus and also for the reference stimulus (i.e. reference white). The interface converts this data into tristimulus values of the form used by the selected color appearance model (which could, for example, be X, Y, Z), while also carrying out the needed transform, which is based on an estimate of the subject's likely previous long-term adaptations to their unique cone fundamentals. The selected standard color appearance model then provides color appearance data that is an estimate of the color appearance of the test stimulus, for that individual subject. This information has the advantage of being interpretable within that model's well-known color space. The adaptive front end is based on the fact that, for any selected input SPD and the subject's unique color matching functions, there can be many different SPDs that are metameric for that individual. Since observer-to-observer color perception differences are minimized for spectrally smooth SPDs, smooth metamers predict color appearances reasonably accurately.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it