Abdul Muntakim Rafi
I am a Ph.D. candidate in Biomedical Engineering at the University of British Columbia. I completed my Bachelor’s degree in Electrical and Electronic Engineering at Bangladesh University of Engineering & Technology and later pursued my Master’s in Electrical and Computer Engineering at University of Windsor.
My current research concentrates on employing machine learning to design cis-regulatory models, develop methods to interpret them, and explore ways to enhance their performance, contributing to a quantitative comprehension of cis-regulatory logic. I have expertise in various programming languages and machine learning libraries, publications in top-tier ML conferences, and experience of working as a machine learning engineer in software companies and startups. I have supervised multiple Co-op students in the de Boer lab and am always on the lookout for motivated undergrads/high schoolers to join my research endeavors.
Email  | 
CV  | 
Google Scholar
 | 
LinkedIn  | 
ResearchGate  | 
Github  | 
CV of Failures
|
|
|
A community effort to optimize sequence-based deep learning models of gene regulation
[Data] [Code]
Abdul Muntakim Rafi, Daria Nogina, Dmitry Penzar, Dohoon Lee, Danyeong Lee, Nayeon Kim, Sangyeup Kim, Dohyeon Kim, Yeojin Shin, Il-Youp Kwak, Georgy Meshcheryakov, Andrey Lando, Arsenii Zinkevich, Byeong-Chan Kim, Juhyun Lee, Taein Kang, Eeshit Dhaval Vaishnav, Payman Yadollahpour, Random Promoter DREAM Challenge Consortiumstrong>, Sun Kim, Jake Albrecht, Aviv Regev, Wuming Gong, Ivan V. Kulakovskiy, Pablo Meyer, Carl de Boer
Random Promoter DREAM Challenge 2022  
Neural networks have proven to be an immensely powerful tool in predicting functional genomic regions, in particular with many recent successes in deciphering gene regulatory logic. However, how model architecture and training strategy choices affect model performance has not been systematically evaluated for genomics models. To address this gap, we held a DREAM Challenge ...
|
|
Application of DenseNet in Camera Model Identification and Post-processing Detection
Abdul Muntakim Rafi, Uday Kamal, Rakibul Hoque, Abid Abrar, Sowmitra Das, Robert Laganiere, Md Kamrul Hasan
CVPR 2019  
[Code]
Camera model identification has earned paramount importance in the field of image forensics with an upsurge of digitally altered images which are constantly being shared through websites, media, and social applications. But, the task of identification becomes quite challenging if metadata are absent from the image and/or if the image has been postprocessed. In this paper, we present ...
|
|
RemNet: remnant convolutional neural network for camera model identification
Abdul Muntakim Rafi, Thamidul Islam Tonmoy, Uday Kamal, Jonathan Wu, Md Kamrul Hasan
Neural Computing and Applications  
[Code]
In this paper, a novel convolutional neural network (CNN) architecture is proposed for CMI with emphasis given on the preprocessing task considered to be inevitable for removing the scene content that heavily obscures the camera model fingerprints. Unlike the conventional approaches where fixed filters are used for preprocessing, the proposed remnant blocks, when coupled with a classification block and trained end-to-end minimizing the classification loss, learn to suppress the unnecessary image contents dynamically ...
|
|
Lung cancer tumor region segmentation using recurrent 3d-denseunet
Uday Kamal, Abdul Muntakim Rafi, Rakibul Hoque, Jonathan Wu, Md Kamrul Hasan
MICCAI 2020  
[Code]
In this paper, we present Recurrent 3D-DenseUNet, a novel deep learning based architecture for volumetric lung tumor segmentation from CT scans. The proposed architecture consists of a 3D encoder block that learns to extract fine-grained spatial and coarse-grained temporal features, a recurrent block of multiple Convolutional Long Short-Term Memory (ConvLSTM) layers to extract fine-grained spatio-temporal ...
|
|
Image-based Bengali Sign Language Alphabet Recognition for Deaf and Dumb Community
Abdul Muntakim Rafi, Nowshin Nawal, Nur Sultan Nazar Bayev, Lusain Nima, Celia Shahnaz, Shaikh Anowarul Fattah
IEEE GHTC 2019  
We have collected in total 12581 different hand signs for the
38 BdSL alphabets in collaboration with the National Federation
of the Deaf. We propose a VGG19 based convolutional neural
network for the recognition of 38 classes and achieve an overall
test accuracy of 89.6% ...
|
|
Mitacs Accelerate Intern, Lanner Electronics Inc.
Joined Lanner through the Mitacs Accelerate, which is Canada's premiere research internship program. Here, I worked
on efficient inference of different AI-driven applications in edge devices.
|
|
Deep Learning Engineer, IFIVEO.
Joined ifiveo through the Mitacs Accelerate. Here, my task has been to perform activity recognition in order to measure
and improve manufacturing floor production processes using deep learning based vision systems. I have collected data from
manufacturing floors, supervised the annotation process, and deployed deep learning models using Amazon Sagemaker.
|
|
Machine Learning Engineer, REVE Systems Ltd.
Worked on designing a real-time Sign2Text translator for Bangla Sign Language.
|
|