Depth discrimination from occlusions in 3D clutter scenes
Bibliographic record
Abstract
Objects such as trees, shrubs, and tall grass typically consist of thousands of small surfaces that are distributed randomly over a 3D volume. Despite the common occurrence of such 3D clutter in natural scenes, relatively little is known about how well humans can perceive depths of surfaces within such 3D clutter. Previous studies have concentrated on motion parallax and binocular disparity cues and have asked questions such as how many discrete depth planes can be perceived, and what is the depth-to-width ratio. However, these studies are incomplete because they have ignored occlusions which are omnipresent in such scenes. Here we present a depth discrimination experiment that examines occlusion cues directly. The task is to discriminate the depths of two red target surfaces in a 3D field of random gray distractors. The experiment uses an Oculus Rift DK2 display which allows us to control motion parallax and binocular disparity cues. The clutter itself provides two occlusion-based depth cues. The first is a 'visibility cue', namely, the target that is less visible is more likely to be deeper within the clutter [Langer and Mannan, JOSA 2012]. The second is a 'context cue', namely the target with the deepest occluder is itself more likely to be deeper. We define scene classes with all four combinations of visibility and occlusion cues and use staircases to measure depth discrimination thresholds. Our results show that observers use both visibility and context cues to perform the task, even when stereo and motion parallax cues are also present. To our knowledge, this is the first experiment to examine depth from occlusions in 3D clutter and to identify the important role played by visibility and context cues in solving this natural problem. Meeting abstract presented at VSS 2016
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 | 0.003 |
| 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 | 0.001 | 0.001 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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".