News
- [Oct 23] Our paper got accepted in the NeurIPS 2023 M3L Workshop 2023
- [Oct 23] I presented our work at ICIP 2023 in Malaysia
- [Jun 23] Our work got accepted in ICIP 2023
- [Feb 23] Our paper is published in Multimedia Tools and Applications
- [Jan 22] Our paper got accepted in ICASSP 2023
|
Research
AI is my passion. I am interested in deep learning and its applications. I have the experience of using DL architectures for solving problems in a variety of domains from medical scans to 5G networks. My research covers different aspects of DL from high-level learning paradigms such as continual learning to transistor-level hardware implementation.
Citations: 0 | H-Index: 0 | i10-Index: 0
|
|
Continual Learning for Out-of-Distribution Generalization in Pedestrian Detection
Mahdiyar Molahasani, Ali Etemad, Michael Greenspan
arXiv/Code
International Conference of Image Processing (ICIP), 2023
A
The study introduces a continual learning approach for pedestrian detection that can adapt to shifts in data distribution, addressing a common issue in current models. By modifying Elastic Weight Consolidation in an object detection network, the method maintains performance across different datasets, preventing catastrophic forgetting demonstrating significant improvements in miss rate reduction on the CrowdHuman and CityPersons datasets when compared to standard fine-tuning. |
|
Multi-scale Multi-task Crowd Counting
Mohsen Zand, Haleh Damirchi, Andrew Farley, Mahdiyar Molahasani, Michael Greenspan, Ali Etemad
arXiv/Code
International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2022
A multi-scale crowd counting and localization platform is proposed in this work. This novel architecture alongside with the multi-scale multi-task loss function have demonstrated promising performance.
|
|
MSG-Caps GAN for Face Super-Resolution
Mahdiyar Molahasani, Seok-bum Ko
Conference/Code
International Conference on Electronics, Information, and Communication (ICEIC), 2020
Multimedia Tools and Applications, 2020
We proposed the first Multi-scale gradient capsule GAN and utlized it for face super-resolution. This model outperformed state-of-the-art face super-resolution models.
|
|
COVID-CXNet: Detecting COVID-19 in Frontal Chest X-ray Images
Arman Haghanifar, Mahdiyar Molahasani, Younhee Choi, S Deivalakshmi, Seok-bum Ko
arXiv/Code
Multimedia Tools and Applications, 2021
In this work, the largest publicly available dataset for COVID19 is collected and a powerful COVID19 detection model based on CheXNet is proposed. This model can detect COVID19 accurately using meaningful features
|
|
High-scale Prostate MRI Super-Resolution with MSG-CapsGAN
Mahdiyar Molahasani, Younhee Choi, S Deivalakshmi, Seok-bum Ko
Multimedia Tools and Applications, 2021
One of the first attempts for high-scale super-resolution (8x) in biomedical domain. MSG-CapsGAN shows promising results in the medical domain as well.
|
|
Automated Tooth Extrcation and Caries Detetcion
Arman Haghanifar, Mahdiyar Molahasani, Seok-bum Ko
arXiv/Conference/Code (extraction)/Code (detection)
IEEE International Symposium on Circuits and Systems (ISCAS), 2020
Multimedia Tools and Applications, (Under review)
A fully automated tooth extcartion model is implemented using genetic algorithm. A multi-feature extraction model with capsule classifier is developed for caries detection.
|
|
Hybrid CMOS/Memristor Crossbar implementation of Recurrent Neural Networks
Mahdiyar Molahasani, Jafar Shamsi, S. B. Shokouhi, Seok-bum Ko
Hopfield/BAM/Code
Analog Integrated Circuits and Signal Processing, 2021
Microelectronics Journal, 2020
An efficient and scalable transistor level implementation of two different recurrent neural networks is proposed using memristor crossbar array.
|
|
Anomaly Prediction in 5G Network
Ramin Sharifi, Mahdiyar Molahasani, Vahid Tabataba Vakili
IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM), 2019
Using LSTM network for user activity prediction in 5G network. The proposed model is capable of anomaly prediction one hour ahead.
|
|
Erosion Detection in Hydraulic Tubes and Hoses Using GRU
Elnaz Etminan, Mahdiyar Molahasani, Seok-bum Ko, Travis Wiens
Fluid Power Systems Technology, American Society of Mechanical Engineers, 2021
The properties of the eroded area in the pipe is extracted from the pressure responce using GRU network. This work is the first erison dtection system using deep learning.
|
|