Search Results for author: Shohreh Kasaei

Found 25 papers, 6 papers with code

AICSD: Adaptive Inter-Class Similarity Distillation for Semantic Segmentation

1 code implementation8 Aug 2023 Amir M. Mansourian, Rozhan Ahmadi, Shohreh Kasaei

With inference time being a crucial factor, particularly in dense prediction tasks such as semantic segmentation, knowledge distillation has emerged as a successful technique for improving the accuracy of lightweight student networks.

Knowledge Distillation Semantic Segmentation

Trainable Loss Weights in Super-Resolution

1 code implementation25 Jan 2023 Arash Chaichi Mellatshahi, Shohreh Kasaei

In this article, a new weighting method for pixel-wise loss is proposed.

Image Super-Resolution

Model-Free Prediction of Adversarial Drop Points in 3D Point Clouds

no code implementations19 Oct 2022 Hanieh Naderi, Chinthaka Dinesh, Ivan V. Bajic, Shohreh Kasaei

To this end, we define 14 point cloud features and use multiple linear regression to examine whether these features can be used for model-free adversarial point prediction, and which combination of features is best suited for this purpose.

Decision Making

LPF-Defense: 3D Adversarial Defense based on Frequency Analysis

2 code implementations23 Feb 2022 Hanieh Naderi, Kimia Noorbakhsh, Arian Etemadi, Shohreh Kasaei

Although 3D point cloud classification has recently been widely deployed in different application scenarios, it is still very vulnerable to adversarial attacks.

3D Point Cloud Classification Adversarial Defense +1

Information-Theoretic Analysis of Minimax Excess Risk

no code implementations15 Feb 2022 Hassan Hafez-Kolahi, Behrad Moniri, Shohreh Kasaei

We consider the frequentist problem of minimax excess risk as a zero-sum game between the algorithm designer and the world.

Learning Theory

Adversarial Attack by Limited Point Cloud Surface Modifications

no code implementations7 Oct 2021 Atrin Arya, Hanieh Naderi, Shohreh Kasaei

The obtained results show that it can perform successful attacks and achieve state-of-the-art results by only a limited number of point modifications while preserving the appearance of the point cloud.

Adversarial Attack Point Cloud Classification +1

CHASE: Robust Visual Tracking via Cell-Level Differentiable Neural Architecture Search

1 code implementation2 Jul 2021 Seyed Mojtaba Marvasti-Zadeh, Javad Khaghani, Li Cheng, Hossein Ghanei-Yakhdan, Shohreh Kasaei

A strong visual object tracker nowadays relies on its well-crafted modules, which typically consist of manually-designed network architectures to deliver high-quality tracking results.

Neural Architecture Search Semantic Segmentation +1

Rate-Distortion Analysis of Minimum Excess Risk in Bayesian Learning

no code implementations10 May 2021 Hassan Hafez-Kolahi, Behrad Moniri, Shohreh Kasaei, Mahdieh Soleymani Baghshah

For the upper bound, the optimization is further constrained to use $R$ bits from the training set, a setting which relates MER to information-theoretic bounds on the generalization gap in frequentist learning.

Generating Unrestricted Adversarial Examples via Three Parameters

no code implementations13 Mar 2021 Hanieh Naderi, Leili Goli, Shohreh Kasaei

It also reduces the model accuracy by an average of 73% on six datasets MNIST, FMNIST, SVHN, CIFAR10, CIFAR100, and ImageNet.

Adversarial Attack

Be Your Own Best Competitor! Multi-Branched Adversarial Knowledge Transfer

no code implementations9 Oct 2020 Mahdi Ghorbani, Fahimeh Fooladgar, Shohreh Kasaei

The proposed method has been devoted to both lightweight image classification and encoder-decoder architectures to boost the performance of small and compact models without incurring extra computational overhead at the inference process.

Image Classification Knowledge Distillation +2

Adaptive Exploitation of Pre-trained Deep Convolutional Neural Networks for Robust Visual Tracking

no code implementations29 Aug 2020 Seyed Mojtaba Marvasti-Zadeh, Hossein Ghanei-Yakhdan, Shohreh Kasaei

Third, the generalization of the proposed method is validated on various tracking datasets as well as CNN models with similar architectures.

Visual Tracking

COMET: Context-Aware IoU-Guided Network for Small Object Tracking

no code implementations4 Jun 2020 Seyed Mojtaba Marvasti-Zadeh, Javad Khaghani, Hossein Ghanei-Yakhdan, Shohreh Kasaei, Li Cheng

To address this problem, we introduce a context-aware IoU-guided tracker (COMET) that exploits a multitask two-stream network and an offline reference proposal generation strategy.

Object Tracking

Effective Fusion of Deep Multitasking Representations for Robust Visual Tracking

no code implementations3 Apr 2020 Seyed Mojtaba Marvasti-Zadeh, Hossein Ghanei-Yakhdan, Shohreh Kasaei, Kamal Nasrollahi, Thomas B. Moeslund

Then, the proposed method extracts deep semantic information from a fully convolutional FEN and fuses it with the best ResNet-based feature maps to strengthen the target representation in the learning process of continuous convolution filters.

Semantic Segmentation Visual Object Tracking +1

Lightweight Residual Densely Connected Convolutional Neural Network

1 code implementation2 Jan 2020 Fahimeh Fooladgar, Shohreh Kasaei

Extremely efficient convolutional neural network architectures are one of the most important requirements for limited-resource devices (such as embedded and mobile devices).

Pointwise Attention-Based Atrous Convolutional Neural Networks

no code implementations27 Dec 2019 Mobina Mahdavi, Fahimeh Fooladgar, Shohreh Kasaei

With the rapid progress of deep convolutional neural networks, in almost all robotic applications, the availability of 3D point clouds improves the accuracy of 3D semantic segmentation methods.

3D Semantic Segmentation Segmentation

Deep Learning for Visual Tracking: A Comprehensive Survey

1 code implementation2 Dec 2019 Seyed Mojtaba Marvasti-Zadeh, Li Cheng, Hossein Ghanei-Yakhdan, Shohreh Kasaei

Second, popular visual tracking benchmarks and their respective properties are compared, and their evaluation metrics are summarized.

Visual Tracking

Do Compressed Representations Generalize Better?

no code implementations20 Sep 2019 Hassan Hafez-Kolahi, Shohreh Kasaei, Mahdiyeh Soleymani-Baghshah

In this paper, the constraint on the entropy $H(X)$ of the input variable $X$ is studied as a simplicity assumption.

Information Bottleneck and its Applications in Deep Learning

no code implementations7 Apr 2019 Hassan Hafez-Kolahi, Shohreh Kasaei

Information Theory (IT) has been used in Machine Learning (ML) from early days of this field.

A Novel Boundary Matching Algorithm for Video Temporal Error Concealment

no code implementations25 Oct 2016 Seyed Mojtaba Marvasti-Zadeh, Hossein Ghanei-Yakhdan, Shohreh Kasaei

It then uses a classic boundary matching criterion or the proposed boundary matching criterion adaptively to identify matching distortion in each boundary of candidate MB.


A Bayesian Framework for Sparse Representation-Based 3D Human Pose Estimation

no code implementations29 Nov 2014 Behnam Babagholami-Mohamadabadi, Amin Jourabloo, Ali Zarghami, Shohreh Kasaei

Our Bayesian framework estimates a posterior distribution for the sparse codes and the dictionaries from labeled training data.

3D Human Pose Estimation

Cannot find the paper you are looking for? You can Submit a new open access paper.