Using high dimensional indexes to support relevance feedback based interactive images retrieval
Bibliographic record
Abstract
Image retrieval has found more and more applications. Due to the well recognized semantic gap problem, the accuracy and the recall of image similarity search are often still low. As an effective method to improve the quality of image retrieval, the relevance feedback approach actively applies users ’ feedback to refine the search. As searching a large image database is often costly, to improve the efficiency, high dimensional indexes may help. However, many existing database indexes are not adaptive to updates of distance measures caused by users ’ feedback. In this paper, we propose a demo to illustrate the relevance feedback based interactive images retrieval procedure, and examine the effectiveness and the efficiency of various indexes. Particularly, audience can interactively investigate the effect of updated distance measures on the data space where the images are supposed to be indexed, and on the distributions of the similar images in the indexes. We also introduce our new B +-tree-like index method based on cluster splitting and iDistance. 1. BACKGROUND Image retrieval is important in many applications. Typically, in a similarity search, a user wants to search for images that are similar to a given query image. However, due to the well recognized semantic gap problem [1], the accuracy and the recall of image similarity search are often still low. As an effective method to improve the quality of image retrieval, the relevance feedback approach [13] actively applies users ’ feedback to refine the search. In the first round, a
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 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.005 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 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".