A gaze-based study for investigating the perception of visual realism in simulated scenes
Bibliographic record
Abstract
Visual realism has been a major objective of computer graphics since the inception of the field. However, the perception of visual realism is not a well-understood process and is usually attributed to a combination of visual cues and image features that are difficult to define or measure. For highly complex images, the problem is even more involved. The purpose of this paper is to present a study based on eye tracking for investigating the perception of visual realism of static images with different visual qualities. The eye-fixation clusters helped to define salient image features corresponding to 3D surface details and light transfer properties that attract observers' attention. This enabled the definition and categorization of image attributes affecting the perception of photorealism. The dynamics of the visual behavior of different observer groups were examined by analyzing saccadic eye movements. We also demonstrated how the different image categories used in the experiments were perceived with varying degrees of visual realism. The results presented can be used as a basis for investigating the impact of individual image features on the perception of visual realism. This study suggests that post-recall or simple abstraction of visual experience is not accurate and the use of eye tracking provides an effective way of determining relevant features that affect visual realism, thus allowing for improved rendering techniques that target these features.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 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 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.003 | 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 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".