no code implementations • ECCV 2020 • Fariborz Taherkhani, Ali Dabouei, Sobhan Soleymani, Jeremy Dawson, Nasser M. Nasrabadi
Semi-Supervised Learning (SSL) based on Convolutional Neural Networks (CNNs) have recently been proven as powerful tools for standard tasks such as image classification when there is not a sufficient amount of labeled data available during the training.
1 code implementation • 25 Mar 2024 • Mohammad Saeed Ebrahimi Saadabadi, Ali Dabouei, Sahar Rahimi Malakshan, Nasser M. Nasrabad
Aiming to enhance the utilization of metric space by the parametric softmax classifier, recent studies suggest replacing it with a non-parametric alternative.
no code implementations • 10 Dec 2023 • Zaber Ibn Abdul Hakim, Najibul Haque Sarker, Rahul Pratap Singh, Bishmoy Paul, Ali Dabouei, Min Xu
Orthogonal to the previous approaches to this limitation, we postulate that understanding the significance of the sentence components according to the target task can potentially enhance the performance of the models.
no code implementations • 27 Sep 2023 • Amol S. Joshi, Ali Dabouei, Nasser Nasrabadi, Jeremy Dawson
Limited data availability is a challenging problem in the latent fingerprint domain.
no code implementations • 2 Sep 2022 • Ali Dabouei, Fariborz Taherkhani, Sobhan Soleymani, Nasser M. Nasrabadi
This phenomenon hinders the outer optimization in AT since the convergence rate of MSGD is highly dependent on the variance of the gradients.
no code implementations • 10 Jun 2022 • Paria Jeihouni, Omid Dehzangi, Annahita Amireskandari, Ali Dabouei, Ali Rezai, Nasser M. Nasrabadi
Our ablation study results on the WVU-OCT data-set in five-fold cross-validation (5-CV) suggest that the proposed MultiSDGAN with a serial attention module provides the most competitive performance, and guiding the spatial attention feature maps by binary masks further improves the performance in our proposed network.
no code implementations • 10 Dec 2021 • Sobhan Soleymani, Ali Dabouei, Fariborz Taherkhani, Seyed Mehdi Iranmanesh, Jeremy Dawson, Nasser M. Nasrabadi
The first loss assures that the representations of modalities for a class have comparable magnitudes to provide a better quality estimation, while the multimodal representations of different classes are distributed to achieve maximum discrimination in the embedding space.
no code implementations • 29 Jul 2021 • Fariborz Taherkhani, Ali Dabouei, Sobhan Soleymani, Jeremy Dawson, Nasser M. Nasrabadi
In this paper, we propose a new method to leverage features from human attributes for person ReID.
no code implementations • 29 Jul 2021 • Fariborz Taherkhani, Ali Dabouei, Sobhan Soleymani, Jeremy Dawson, Nasser M. Nasrabadi
To address this problem, we use a new Multi-Task Learning (MTL) paradigm in which a facial attribute predictor uses the knowledge of other related attributes to obtain a better generalization performance.
no code implementations • 21 Jun 2021 • Amol S. Joshi, Ali Dabouei, Jeremy Dawson, Nasser M. Nasrabadi
It is added to generate ridge maps to ensure that fingerprint information and minutiae are preserved in the deblurring process and prevent the model from generating erroneous minutiae.
no code implementations • CVPR 2021 • Fariborz Taherkhani, Ali Dabouei, Sobhan Soleymani, Jeremy Dawson, Nasser M. Nasrabadi
The goal is to use Wasserstein metric to provide pseudo labels for the unlabeled images to train a Convolutional Neural Networks (CNN) in a Semi-Supervised Learning (SSL) manner for the classification task.
no code implementations • 10 Dec 2020 • Syeda Nyma Ferdous, Ali Dabouei, Jeremy Dawson, Nasser M Nasrabadi
High recognition accuracy of the synthesized samples that is close to the accuracy achieved using the original high-resolution images validate the effectiveness of our proposed model.
no code implementations • 4 Dec 2020 • Fariborz Taherkhani, Hadi Kazemi, Ali Dabouei, Jeremy Dawson, Nasser M. Nasrabadi
Semi-Supervised Learning (SSL) approaches have been an influential framework for the usage of unlabeled data when there is not a sufficient amount of labeled data available over the course of training.
no code implementations • 2 Dec 2020 • Sobhan Soleymani, Ali Dabouei, Fariborz Taherkhani, Jeremy Dawson, Nasser M. Nasrabadi
The network is trained by triplets of face images, in which the intermediate image inherits the landmarks from one image and the appearance from the other image.
no code implementations • 2 Dec 2020 • Sobhan Soleymani, Baaria Chaudhary, Ali Dabouei, Jeremy Dawson, Nasser M. Nasrabadi
Although biometric facial recognition systems are fast becoming part of security applications, these systems are still vulnerable to morphing attacks, in which a facial reference image can be verified as two or more separate identities.
no code implementations • 4 Sep 2020 • Seyed Mehdi Iranmanesh, Ali Dabouei, Nasser M. Nasrabadi
We present a novel framework to exploit privileged information for recognition which is provided only during the training phase.
no code implementations • 28 Jul 2020 • Saba Heidari Gheshlaghi, Omid Dehzangi, Ali Dabouei, Annahita Amireskandari, Ali Rezai, Nasser M. Nasrabadi
We incorporate the Unet architecture in the NAS framework as its backbone for the segmentation of the retinal layers in our collected and pre-processed OCT image dataset.
no code implementations • CVPR 2020 • Ali Dabouei, Sobhan Soleymani, Fariborz Taherkhani, Jeremy Dawson, Nasser M. Nasrabadi
Recently, ensemble models have demonstrated empirical capabilities to alleviate the adversarial vulnerability.
1 code implementation • 20 Apr 2020 • Uche Osahor, Hadi Kazemi, Ali Dabouei, Nasser Nasrabadi
We incorporate a hybrid discriminator which performs attribute classification of multiple target attributes, a quality guided encoder that minimizes the perceptual dissimilarity of the latent space embedding of the synthesized and real image at different layers in the network and an identity preserving network that maintains the identity of the synthesised image throughout the training process.
2 code implementations • CVPR 2021 • Ali Dabouei, Sobhan Soleymani, Fariborz Taherkhani, Nasser M. Nasrabadi
On the distillation task, solely classifying images mixed using the teacher's knowledge achieves comparable performance to the state-of-the-art distillation methods.
no code implementations • 13 Jan 2020 • Ali Dabouei, Fariborz Taherkhani, Sobhan Soleymani, Jeremy Dawson, Nasser M. Nasrabadi
We demonstrate that the proposed approach enhances the performance of deep face recognition models by assisting the training process in two ways.
no code implementations • 7 Jan 2020 • Seyed Mehdi Iranmanesh, Ali Dabouei, Sobhan Soleymani, Hadi Kazemi, Nasser M. Nasrabadi
In this work, we present a practical approach to the problem of facial landmark detection.
1 code implementation • 8 Oct 2019 • Ali Dabouei, Sobhan Soleymani, Fariborz Taherkhani, Jeremy Dawson, Nasser M. Nasrabadi
Deep neural networks are susceptible to adversarial manipulations in the input domain.
no code implementations • 8 Aug 2019 • Sobhan Soleymani, Ali Dabouei, Jeremy Dawson, Nasser M. Nasrabadi
Deep neural networks have presented impressive performance in biometric applications.
no code implementations • 21 Jun 2019 • Sobhan Soleymani, Ali Dabouei, Jeremy Dawson, Nasser M. Nasrabadi
Therefore, to compensate for this shortcoming, we propose to train a deep auto-encoder surrogate network to mimic the conventional iris code generation procedure.
1 code implementation • 24 Sep 2018 • Ali Dabouei, Sobhan Soleymani, Jeremy Dawson, Nasser M. Nasrabadi
The state-of-the-art performance of deep learning algorithms has led to a considerable increase in the utilization of machine learning in security-sensitive and critical applications.
no code implementations • 31 Jul 2018 • Sobhan Soleymani, Ali Dabouei, Seyed Mehdi Iranmanesh, Hadi Kazemi, Jeremy Dawson, Nasser M. Nasrabadi
In this paper a novel cross-device text-independent speaker verification architecture is proposed.
no code implementations • 31 Jul 2018 • Seyed Mehdi Iranmanesh, Hadi Kazemi, Sobhan Soleymani, Ali Dabouei, Nasser M. Nasrabadi
The proposed Attribute-Assisted Deep Con- volutional Neural Network (AADCNN) method exploits the facial attributes and leverages the loss functions from the facial attributes identification and face verification tasks in order to learn rich discriminative features in a common em- bedding subspace.
no code implementations • 31 Jul 2018 • Ali Dabouei, Sobhan Soleymani, Hadi Kazemi, Seyed Mehdi Iranmanesh, Jeremy Dawson, Nasser M. Nasrabadi
We achieved the rank-10 accuracy of 88. 02\% on the IIIT-Delhi latent fingerprint database for the task of latent-to-latent matching and rank-50 accuracy of 70. 89\% on the IIIT-Delhi MOLF database for the task of latent-to-sensor matching.
no code implementations • 3 Jul 2018 • Sobhan Soleymani, Ali Dabouei, Hadi Kazemi, Jeremy Dawson, Nasser M. Nasrabadi
Multiple features are extracted at several different convolutional layers from each modality-specific CNN for joint feature fusion, optimization, and classification.
no code implementations • 9 Apr 2018 • Hadi Kazemi, Sobhan Soleymani, Ali Dabouei, Mehdi Iranmanesh, Nasser M. Nasrabadi
Specifically, an attribute-centered loss is proposed which learns several distinct centers, in a shared embedding space, for photos and sketches with different combinations of attributes.
no code implementations • 4 Jan 2018 • Seyed Mehdi Iranmanesh, Ali Dabouei, Hadi Kazemi, Nasser M. Nasrabadi
we propose a coupled deep neural network architecture which leverages relatively large visible and thermal datasets to overcome the problem of overfitting and eventually we train it by a polarimetric thermal face dataset which is the first of its kind.
no code implementations • 3 Jan 2018 • Ali Dabouei, Hadi Kazemi, Seyed Mehdi Iranmanesh, Jeremi Dawson, Nasser M. Nasrabadi
Elastic distortion of fingerprints has a negative effect on the performance of fingerprint recognition systems.