Minimalistic CNN-based ensemble model for gender prediction from face images

Antipov, Grigory; Berrani, Sid-Ahmed; Dugelay, Jean‐Luc

doi:10.1016/j.patrec.2015.11.011

Cited by 100 publications

(68 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In fact, previous state-of-the-art for cross-dataset with LFW has already reported 97%, but that was achieved training with 400,000 samples [11] or four millions [10] Here that accuracy is beaten, reaching 98%, and that is done with just a 7% of the training samples used by Antipov et al When training with MORPH GC rates are lower, 88%, but significantly better that recent reported results that reached 76% [4]. On the other side, GROUPS present larger difficulties, being extremely complex if training with MORPH, just 67%, and easier training with LFW.…”

Section: In-and Cross-database Results In Full Datasetsmentioning

confidence: 96%

“…CNN [53] have lately achieved relevant results in many Computer Vision problems as image classification [54]. In this sense, some authors have started to evaluate them in GC, with some results reported for LFW and GROUPS [6,11]. We have adopted the CNN design proposed by [55] with three convolutional layers and two fully connected layers, trained with HS pattern (159 × 155 pixels), see section 2.…”

Section: In-and Cross-database Results In Full Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Castrilln-Santana

Lorenzo-Navarro

Ramn-Balmaseda

2017

Image and Vision Computing

View full text Add to dashboard Cite

Gender classification (GC) has achieved high accuracy in different experimental evaluations based mostly on inner facial details. However, these results do not generalize well in unrestricted datasets and particularly in cross-database experiments, where the performance drops drastically. In this paper, we analyze the state-of-the-art GC accuracy on three large datasets: MORPH, LFW and GROUPS. We discuss their respective difficulties and bias, concluding that the most challenging and wildest complexity is present in GROUPS. This dataset covers hard conditions such as low resolution imagery and cluttered background. Firstly, we analyze in depth the performance of different descriptors extracted from the face and its local context on this dataset. Selecting the bests and studying their most suitable combination allows us to design a solution that beats any previously published results for GROUPS with the Dago's protocol, reaching an accuracy over 94.2%, reducing the gap with other simpler datasets. The chosen solution based on local descriptors is later evaluated in a cross-database scenario with the three mentioned datasets, and full dataset 5-fold cross validation. The achieved results are compared with a Convolutional Neural Network approach, achieving rather similar marks. Finally, a solution is proposed combining both focuses, exhibiting great complementarity, boosting GC performance to beat previously published results in GC both cross-database, and full in-database evaluations.

show abstract

Section: In-and Cross-database Results In Full Datasetsmentioning

confidence: 96%

Section: In-and Cross-database Results In Full Datasetsmentioning

confidence: 99%

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Castrilln-Santana

Lorenzo-Navarro

Ramn-Balmaseda

2017

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

“…GilNet model, proposed in [LH15], is trained with Adience dataset [EEH14] from scratch for gender estimation. In [ABD16], a CNN ensemble model is proposed for running efficiently in embedded devices. Experiments of this method are conducted on LFW [Hu07] dataset and state-of-the-art results are obtained according to [ABD16].…”

Section: Gender Classificationmentioning

confidence: 99%

How Transferable Are CNN-Based Features for Age and Gender Classification?

Özbulak

Aytar

Ekenel

2016

2016 International Conference of the Biometrics Special Interest Group (BIOSIG)

View full text Add to dashboard Cite

Age and gender are complementary soft biometric traits for face recognition. Successful estimation of age and gender from facial images taken under real-world conditions can contribute improving the identification results in the wild. In this study, in order to achieve robust age and gender classification in the wild, we have benefited from Deep Convolutional Neural Networks based representation. We have explored transferability of existing deep convolutional neural network (CNN) models for age and gender classification. The generic AlexNet-like architecture and domain specific VGG-Face CNN model are employed and fine-tuned with the Adience dataset prepared for age and gender classification in uncontrolled environments. In addition, task specific GilNet CNN model has also been utilized and used as a baseline method in order to compare with transferred models. Experimental results show that both transferred deep CNN models outperform the GilNet CNN model, which is the state-of-the-art age and gender classification approach on the Adience dataset, by an absolute increase of 7% and 4.5% in accuracy, respectively. This outcome indicates that transferring a deep CNN model can provide better classification performance than a task specific CNN model, which has a limited number of layers and trained from scratch using a limited amount of data as in the case of GilNet. Domain specific VGG-Face CNN model has been found to be more useful and provided better performance for both age and gender classification tasks, when compared with generic AlexNet-like model, which shows that transfering from a closer domain is more useful.

show abstract

“…Jia and Cristianini [28] trained with four million images, achieving 96.9%. More recently Antipov et al [3] reported 97.1% assembling three Convolutional Neural Networks (CNNs).…”

Section: Related Workmentioning

confidence: 99%

“…We mentioned above the work by Antipov et al for LFW [3]. However, there is also an interest in combining CNN outputs and local descriptors for GC [47,36].…”

Section: Related Workmentioning

confidence: 99%

Multi-scale score level fusion of local descriptors for gender classification in the wild

Castrillón-Santana

Lorenzo-Navarro

Ramón-Balmaseda

2016

Multimed Tools Appl

View full text Add to dashboard Cite

The 2015 FRVT gender classification (GC) report evidences the problems that current approaches tackle in situations with large variations in pose, illumination, background and facial expression. The report suggests that both commercial and research solutions are hardly able to reach an accuracy over 90% for The Images of Groups dataset, a proven scenario exhibiting unrestricted or in the wild conditions. In this paper, we focus on this challenging dataset, stepping forward in GC performance by observing: 1) recent literature results combining multiple local descriptors, and 2) the psychophysics evidences of the greater importance of the ocular and mouth areas to solve this task. We therefore make use of holistic and inner facial patches to extract features, that are later combined via a score level fusion strategy. The achieved results support the main information provided by the ocular and the mouth areas. Indeed, the combination of multiscale extracted features increases the overall accuracy to over 94%, reducing notoriously the classification error if compared with tuned holistic and deep learning approaches.

show abstract

Minimalistic CNN-based ensemble model for gender prediction from face images

Cited by 100 publications

References 13 publications

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

Descriptors and regions of interest fusion for in- and cross-database gender classification in the wild

How Transferable Are CNN-Based Features for Age and Gender Classification?

Multi-scale score level fusion of local descriptors for gender classification in the wild

Contact Info

Product

Resources

About