Deep neural network concepts for background subtraction: A systematic review and comparative evaluation
Conventional neural networks have been demonstrated to be a powerful framework for
background subtraction in video acquired by static cameras. Indeed, the well-known Self …
background subtraction in video acquired by static cameras. Indeed, the well-known Self …
Post-training quantization for vision transformer
Recently, transformer has achieved remarkable performance on a variety of computer vision
applications. Compared with mainstream convolutional neural networks, vision transformers …
applications. Compared with mainstream convolutional neural networks, vision transformers …
Proflip: Targeted trojan attack with progressive bit flips
Abstract The security of Deep Neural Networks (DNNs) is of great importance due to their
employment in various safety-critical applications. DNNs are shown to be vulnerable against …
employment in various safety-critical applications. DNNs are shown to be vulnerable against …
Trojvit: Trojan insertion in vision transformers
Abstract Vision Transformers (ViTs) have demonstrated the state-of-the-art performance in
various vision-related tasks. The success of ViTs motivates adversaries to perform backdoor …
various vision-related tasks. The success of ViTs motivates adversaries to perform backdoor …
Selective prototype network for few-shot metal surface defect segmentation
Metal surface defects segmentation is a critical task to make pixel-level predictions about
defects in the industrial production process, which has great significance in improving …
defects in the industrial production process, which has great significance in improving …
Overcoming limitation of dissociation between MD and MI classifications of breast cancer histopathological images through a novel decomposed feature-based …
M Sepahvand, F Abdali-Mohammadi - Computers in Biology and Medicine, 2022 - Elsevier
Magnification-independent (MI) classification is considered a promising method for detecting
the histopathological images of breast cancer. However, it has too many parameters for real …
the histopathological images of breast cancer. However, it has too many parameters for real …
Deep neural network compression through interpretability-based filter pruning
K Yao, F Cao, Y Leung, J Liang - Pattern Recognition, 2021 - Elsevier
This paper proposes a method to compress deep neural networks (DNNs) based on
interpretability. For a trained DNN model, the activation maximization technique is first used …
interpretability. For a trained DNN model, the activation maximization technique is first used …
Trojtext: Test-time invisible textual trojan insertion
In Natural Language Processing (NLP), intelligent neuron models can be susceptible to
textual Trojan attacks. Such attacks occur when Trojan models behave normally for standard …
textual Trojan attacks. Such attacks occur when Trojan models behave normally for standard …
Nonlinear tensor train format for deep neural network compression
Deep neural network (DNN) compression has become a hot topic in the research of deep
learning since the scale of modern DNNs turns into too huge to implement on practical …
learning since the scale of modern DNNs turns into too huge to implement on practical …
Teacher–student knowledge distillation based on decomposed deep feature representation for intelligent mobile applications
M Sepahvand, F Abdali-Mohammadi… - Expert Systems with …, 2022 - Elsevier
According to the recent studies on feature-based knowledge distillation (KD), a student
model will not be able to imitate a teacher's behavior properly if there is a high variance …
model will not be able to imitate a teacher's behavior properly if there is a high variance …