Performance Enhancement For Gender Recognition Using Trainable Bank of Gabor Filters and NCA

Basulaim, Khaled Omer; Dabash, Adeeb Ali

doi:10.1109/icoice48418.2019.9035141

Cited by 3 publications

(3 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Basulaim and Dabash [4] used a bank of Gabor filters with auto configured using parts of facial images, FERET_DATASET is used which contains 474 training and 472 testing face images. A bank of Gabor filters is created and configured for each sub-face which is called COSFIRE (Combination of Shifted Filter Responses) and choose multi-interest prototypes automatically.…”

Section: Related Workmentioning

confidence: 99%

Contribution of Interesting Facial Parts on Gender Recognition System Performance

2021

jecet

View full text Add to dashboard Cite

Recently, the gender recognition system is very important for interaction applications such as consultation questions, legal gender access. There are many research works which used facial information to detect the gender type. One of the most interesting research works is Azzopardi work who used bank of Gabor filters to segment randomly-chosen sub faces. In this research, his proposed model is evaluated on manually-chosen interesting sub-faces such as eye, eyebrow, nose, mouth and mixture of all sub-faces. The observed results showed enhancement in classification accuracy that achieved 95.10% by selecting all the interesting parts of facial image.

show abstract

Section: Related Workmentioning

confidence: 99%

Contribution of Interesting Facial Parts on Gender Recognition System Performance

2021

jecet

View full text Add to dashboard Cite

show abstract

“…LFW+ Antipov et al (2017) 0.993 Antipov et al (2016) 0.971 Jia and Cristianini (2015) 0.968 GF Simanjuntak and Azzopardi (2019) 0.974 Basulaim and Dabash (2019) 0.958 Azzopardi et al (2018b) 0.947 Adience Gurnani et al (2019) 0.918 Dehghan et al (2017) 0.910 Afifi and Abdelhamed (2019) 0.906 MG Azzopardi et al (2018b) 0.915 Carletti et al (2020a) 0.902 Azzopardi et al (2017) 0.802 small architectures on LFW, LFWA, LFW+, GENDER-FERET and GENDER-COLOR-FERET (Del Coco et al 2016); on the other hand, it emerges the evidence that on different test sets the performance can drop below 0.90, especially with cross-dataset evaluations, as happens for Adience and MIVIA-Gender. Only in a few cases the analysis is carried out on more than one benchmark (Azzopardi et al 2018b) (Foggia et al 2019) (Afifi and Abdelhamed 2019) to evaluate the generalization capabilities of the methods and, in most of the papers, the experimentation is performed on images not representing the real conditions.…”

Section: Datasetmentioning

confidence: 99%

“…SVM classifiers trained with COSFIRE filters (Azzopardi et al 2016b) or with a fusion of these features and local SURF descriptors (Azzopardi et al 2018b) obtain 0.936 and 0.947 of accuracy, respectively. Basulaim and Dabash (2019) further improve the performance to 0.958 by using COS-FIRE filters and a cubic SVM. Simanjuntak and Azzopardi (2019) retain the current state of the art performance on this dataset (0.974) with a fusion of CNN and COSFIRE-based features.…”

Section: Introductionmentioning

confidence: 99%

Gender recognition in the wild: a robustness evaluation over corrupted images

Greco

Vento

Vigilante³

2020

J Ambient Intell Human Comput

View full text Add to dashboard Cite

In the era of deep learning, the methods for gender recognition from face images achieve remarkable performance over most of the standard datasets. However, the common experimental analyses do not take into account that the face images given as input to the neural networks are often affected by strong corruptions not always represented in standard datasets. In this paper, we propose an experimental framework for gender recognition “in the wild”. We produce a corrupted version of the popular LFW+ and GENDER-FERET datasets, that we call LFW+C and GENDER-FERET-C, and evaluate the accuracy of nine different network architectures in presence of specific, suitably designed, corruptions; in addition, we perform an experiment on the MIVIA-Gender dataset, recorded in real environments, to analyze the effects of mixed image corruptions happening in the wild. The experimental analysis demonstrates that the robustness of the considered methods can be further improved, since all of them are affected by a performance drop on images collected in the wild or manually corrupted. Starting from the experimental results, we are able to provide useful insights for choosing the best currently available architecture in specific real conditions. The proposed experimental framework, whose code is publicly available, is general enough to be applicable also on different datasets; thus, it can act as a forerunner for future investigations.

show abstract

Fusion of Deep Learning Features for Wild Animal Detection

Dhillon

Verma

2022

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Performance Enhancement For Gender Recognition Using Trainable Bank of Gabor Filters and NCA

Cited by 3 publications

References 12 publications

Contribution of Interesting Facial Parts on Gender Recognition System Performance

Contribution of Interesting Facial Parts on Gender Recognition System Performance

Gender recognition in the wild: a robustness evaluation over corrupted images

Fusion of Deep Learning Features for Wild Animal Detection

Contact Info

Product

Resources

About