High F-score Model for Recognizing Object Visibility in Images with Occluded Objects of Interest
Bibliographic record
Abstract
Article investigates recognition of partially occluded objects of interest.Images from retail store self checkout area often contain products that are covered by a customer's body parts, are placed inside semi-transparent plastic bags, include intensive glare, or some combination of these.In order to categorize objects of interest in images with partially occluded objects, the first step is to decide if an image contains enough information about the object of interest in order to be categorized.The most famous computer vision data sets -such as Imagenet, Canadian Institute for Advanced Research (CIFAR), Digits by National Institute of Standards and Technology (MNIST) -are made of images that contain clearly visible, distinctive objects of interest and are only labelled with binary information about object existence; reduced visibility objects are absent in the mentioned datasets.Such binary visibility labels are not suitable for solving the recognition task of object occlusion level.In this study authors categorize images into [not] containing enough information about objects of interest in order to be categorized.Authors analyze a dataset collected in a real retail store self checkout area where objects of interest are various products.The proposed method uses 6 categories of occlusion variously grouped.Authors received >0.9 F-score in best model separating images into object visible/invisible categories.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".