MétaCan
Menu
Back to cohort

QHash: An efficient hashing algorithm for low-variance image deduplication

2021· article· en· W4285327104 on OpenAlexaff
Xuan Li, Liqiong Chang, Xue Liu

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicAdvanced Steganography and Watermarking Techniques
Canadian institutionsMcGill University
Fundersnot available
KeywordsData deduplicationComputer scienceHash functionVariance (accounting)AlgorithmImage (mathematics)Artificial intelligenceDatabaseComputer security

Abstract

fetched live from OpenAlex

Attributed to the widespread use of general artificial intelligence (AI), large-scale datasets have become a critical component for the success of AI-powered applications. While collecting a larger dataset is desirable in general, studies have revealed that many databases include duplicated images. The accumulation of redundant images will result in excessive resource usage and inefficient cloud storage utilization. To get rid of the duplicates, hashing-based methods have been developed for the problem of image deduplication. However, as demonstrated in our experiments, current approaches fail in dataset with small visual difference, such as medical images. To this end, we propose QHash, which achieves effective image deduplication on the low-variance dataset. QHash leverages Vector Quantized Variational AutoEncoder (VQ-VAE) to learn the data distribution in an unsupervised manner. In addition, hash sequences are implemented using integer-based tensor, enabling distance calculation and deduplication being processed in parallel on either CPU or GPU. Extensive experiments show that QHash outperforms other baseline approaches by at least 50%, while is about 23% memory efficient and 18% deduplication time speedup.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.003
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.003
Threshold uncertainty score0.011

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.003
Meta-epidemiology (narrow)0.0010.000
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0010.001
Science and technology studies0.0010.001
Scholarly communication0.0010.002
Open science0.0010.002
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0030.001

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.

Opus teacher head0.013
GPT teacher head0.275
Teacher spread0.262 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreMethods

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".

Quick stats

Citations0
Published2021
Admission routes1
Has abstractyes

Explore more

Same topicAdvanced Steganography and Watermarking TechniquesFrench-language works237,207