2020
DOI: 10.3390/s20164491
|View full text |Cite
|
Sign up to set email alerts
|

Assessment and Estimation of Face Detection Performance Based on Deep Learning for Forensic Applications

Abstract: Face recognition is a valuable forensic tool for criminal investigators since it certainly helps in identifying individuals in scenarios of criminal activity like fugitives or child sexual abuse. It is, however, a very challenging task as it must be able to handle low-quality images of real world settings and fulfill real time requirements. Deep learning approaches for face detection have proven to be very successful but they require large computation power and processing time. In this work, we evaluate the sp… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
18
0
2

Year Published

2021
2021
2024
2024

Publication Types

Select...
5
4

Relationship

2
7

Authors

Journals

citations
Cited by 31 publications
(20 citation statements)
references
References 58 publications
0
18
0
2
Order By: Relevance
“…Besides, the best compromise between speed and detection was achieved with the DSFD model with an image reduction of 50%. Comparing the different CPUs, the overall best time was achieved by i9-8650HK, which excels at its bus speed 1 . GPU results, however, overcome the CPU ones, and we noticed that the GPU architecture determines the detection speed.…”
Section: Methodsmentioning
confidence: 99%
“…Besides, the best compromise between speed and detection was achieved with the DSFD model with an image reduction of 50%. Comparing the different CPUs, the overall best time was achieved by i9-8650HK, which excels at its bus speed 1 . GPU results, however, overcome the CPU ones, and we noticed that the GPU architecture determines the detection speed.…”
Section: Methodsmentioning
confidence: 99%
“…We selected the keyframes containing at least one face, detected with a confidence score, T , to create the final video summary. As face detector, we chose the Multi-Task Cascade CNN (MTCNN) [23] method due to its high detection speed in comparison to other deep-learning-based detectors [6].…”
Section: Selection Of Keyframes Containing Facesmentioning
confidence: 99%
“…Bu noktadan hareketle literatür araştırmasında özellikle son 5 yılda adli bilişim alanında derin öğrenme algoritmalarının kullanılmasına ilişkin çalışma ve uygulamalar araştırılmıştır. Araştırma neticesinde adli bilişim alanı kapsamında derin öğrenme algoritmalarının yoğun olarak metin verisi [5], ses verisi [6], video [7][8][9] ve resim [3], [10][11][12][13][14][15][16][17][18][19] verileri analizinde kullanıldığı tespit edilmiştir. Söz konusu çalışma alanlarına ek olarak yöntem iyileştirme [20,21], süreç iyileştirme ve otomasyon [22,23], saldırı tespit ve güvenlik log analizi [2], [24][25], derin öğrenme algoritmalarının yaygın olarak kullanıldığı diğer alanlardır.…”
Section: Li̇teratür Araştirmasiunclassified
“…Bu alanda yapılan diğer bir çalışmada hız ve doğruluk açısından derin öğrenme teknolojilerini kullanan üç popüler yüz tanımlama dedektörü farklı veri setleri ile test edilmiştir [8]. Elde edilen sonuçlar, görüntülerin yeniden boyutlandırılarak işleme alınmasının yüz tanıma süreçlerini hızlandırmakla beraber doğruluğunu azalttığını göstermiştir.…”
Section: Li̇teratür Araştirmasiunclassified