MétaCan
Menu
Back to cohort
Record W4399534558 · doi:10.1145/3641032.3641056

Unlocking the Potential of Face Recognition in OpenCV: A Comprehensive Study of Algorithmic Approaches

2023· article· en· W4399534558 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldEngineering
TopicVehicle License Plate Recognition
Canadian institutionsUniversity of Windsor
FundersUniversity of Windsor
KeywordsComputer scienceFacial recognition systemFace (sociological concept)Artificial intelligenceComputer visionFace detectionPattern recognition (psychology)

Abstract

fetched live from OpenAlex

Face recognition technology has become increasingly prevalent in a wide range of industries, including security, monitoring, and biometrics. However, despite this prevalence, achieving accurate and effective face recognition in real-world scenarios remains a challenge. The objective of this research is to examine the algorithmic methods used by OpenCV library for facial recognition and assess their potential to maximize the system's effectiveness. The distinctiveness of this work rests in the comprehensive assessment of both conventional and deep learning-based approaches using OpenCV and Python. Additionally, it involves comparing their performance on a sizable facial dataset, considering factors like speed, accuracy, and precision. The study involves experimentation and testing of three conventional models: Eigenfaces, Fisherfaces, and LBPH. Our findings reveal that these conventional models perform inadequately in situations involving varying lighting conditions, and complex multi-facial contexts while only supporting grayscale images. Thus, we further delved into deep learning models like MTCNN, and pre-trained models like VGG16. While MTCNN exhibited remarkable results with the highest accuracy level, it encountered challenges in scenarios with fluctuating lighting conditions. Whereas as VGG16 yielded comparable outcomes but demanded high-end computational resources. Upon additional experimentation with deep learning models, we found that fine-tuning pre-trained models substantially improved performance on the target dataset, yielding even better results. We concluded that deep learning-based methods can effectively harness OpenCV's facial recognition capabilities, offering an advantage over conventional models. However, it's important to note that the applicability of these models still relies on specific use cases. Our study thoroughly deliberates on the advantages and limitations of each model, enabling the scientific and academic community to make informed decisions, while selecting an appropriate model for distinct use cases. The implications of our study extend to various industries, such as security, surveillance, and biometrics, where precise and effective facial recognition holds paramount importance.

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.301
Threshold uncertainty score0.320

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.080
GPT teacher head0.242
Teacher spread0.163 · 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

Quick stats

Citations0
Published2023
Admission routes2
Has abstractyes

Explore more

Same topicVehicle License Plate RecognitionFrench-language works237,207