Reverse Image Search for Collage: A Novel Local Feature-Based Framework
Bibliographic record
Abstract
Collage, a popular form of visual-content summarization technique is commonly used by internet users and digital artists. Social media usage is a rising trend that significantly affects the increasing demand for collages. The primary source of collage generation is social media, but other sources also generate it. Searching for a required query image in this corpus is a crucial demand and also valuable. The query image can be retrieved using Reverse Image Search (RIS), either in its exact form or with a small variation. Well-known search engines like Google and Yandex have this functionality, but their method has not been made public. In this research, we propose a consolidated framework for reverse image searching for the problem of collage. Essentially, the local features of collage images are extracted by using SIFT, SURF, and ORB algorithms. These features undergo the localization of the region of interest (ROI) process which handles by binning technique. We propose to use the Manhattan distance to calculate the similarity. The proposed model is extensively evaluated on standard databases and is shown to always have good results using SIFT algorithm. The proposed model is entirely generic and attains 90.96%, accuracy using the SIFT algorithm. The proposed approach is also evaluated on flip and scale variant college and achieves a result of 83% and 78% respectively, using SIFT algorithm.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 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".