July 2022 - Present

Department of Computer Science and Engineering

Bangladesh University of Engineering and Technology

Dhaka, Bangladesh

• Working on a project titled ‘Speech to Sign Language: A Deep Learning Approach for Bangla Sign Language Production’ which has been approved and funded by ICTD of the Bangladesh Government.

• Work with deep learning, PyTorch and speech processing

• For more information: Project Overview

April 2022 - Present

Everforth

Tokyo, Japan (Remote)

• Contribution to GrowthPalette, an e-commerce platform jointly developed with ec companies with annual sales of 10 billion yen.

• Work with CakePHP, Vue.js, Docker

• For more information: Everforth, GrowthPalette

Accepted in Neurocomputing- Elsevier on 1 February 2022

• Authors: Md. Ashraful Islam, Mir Mahathir Mohammad, Sarkar Snigdha Sarathi Das, Mohammed Eunus Ali

• Go to Paper

• To the best of our knowledge, this work is the first comprehensive survey of all major deep learning-based POI recommendation works. 

• Our work categorizes and critically analyses the recent POI recommendation works based on different deep learning paradigms and other relevant features. 

• This review can be considered a cookbook for researchers or practitioners working in the area of POI recommendation.

Accepted in 2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition (FG)

Authors: Iftekhar E Mahbub Zeeon, Mir Mahathir Mohammad, Mohammad Abdullah Adnan

Go to Paper

Sign language, a vital communication tool for individuals with hearing impairments worldwide, is prominently utilized within the BangIa-speaking community. BangIa is recognized as the seventh most spoken language globally. However, research in BangIa Sign Language Recognition (SLR) - the process of translating symbols or words from images and videos - has been predominantly confined to controlled environments with limited samples and rudimentary symbol annotations, impeding its application in real-world scenarios. In contrast to previous studies, our research concentrates on SLR. It delves into the relatively unexplored territories of BangIa Sign Language Translation (SLT) and Sign Language Production (SLP), areas that have been largely overlooked due to dataset constraints. We introduce BTVSL, a comprehensive BangIa Sign Language dataset derived from the YouTube series ‘BTV Desh o Jonopoder Khobor'. This dataset, featuring 60 hours of news content in sign language delivered by professionals, represents the largest sentence-level dataset available for BangIa SLT, encompassing a broad spectrum of expressions. Leveraging BTVSL, we evaluated four distinct SLT models, achieving an average BLEU score of 20.42. This result underscores the potential of BTVSL in enhancing the accuracy of sign language translation

February 2022

Convolutional Neural Network, Tensorflow, Kaggle Contest

• The goal of this competition is to accurately identify starfish in real-time by building an object detection model trained on underwater videos of coral reefs. 

• We completed this contest as our CSE 472 Machine Learning Sessional final project submission.

• For more information: Kaggle Specification, Our Demonstration

August 2020

Convolutional Neural Network, Tensorflow, Kaggle Contest

• Identify melanoma in lesion images.

• Reached top 70% ranking

• For more information: My Code, Specification

August 2020

Convolutional Neural Network, Keras, YOLOv3, Kaggle Contest

• Identify wheat heads using image analysis

• Reached top 70% ranking

• For more information: My Code, Specification, My Submission

January 2020 - Present

Node.js, Heroku, Vue.js, React.js, Firebase Hosting, MongoDB, Capacitor.js

• The management website of Badhan, BUET Zone to manage emergency blood donations

• Development of backend, web frontend, hybrid android app, documentation, backup and restoring of database, automated testing of backend API endpoints, implementation of admin console and hosting of the full stack.

• For more information: Github, Website, Android App, Video Presentation, Certificate

March, 2021 - July, 2021

Node.js, Heroku, Vue.js, Firebase Hosting, MySQL, ClearDB Hosting, GCP Storage

• This is an academic platform as our final software development project for the students of the department of CSE of BUET to share and store academic questions, answers and study materials including theses and projects.

• I managed and led a group of six people, administered the system, established and deployed the full stack

• For more information: Backend Repository, Frontend Repository, Website, Design

May 2020

Node.js, Vue.js, Express, MongoDB

• This is a voluntary project duing the times of peak COVID-19

• Benefits you will get on our platform: Participating relief agencies can find relief program information through the map. You can monitor different types of relief programs on the map in different ways through various "filters".

• For more information: frontend repository, backend, Video Demo

October, 2019

PHP, MySQL

• A  social media site for uploading, storing, posting photos.

• This is our database project of Level 2 term 2 at BUET

• You can create an account; post photos with captions and tags; like and comment on any photo

• You may create groups and people can join and see the photos of that group

• A personal photo storage is available for you to store photos

• People, groups and photos can be searched from the search bar

• You can send, accept and reject friend requests

• A user will become a star-user if he/she is frequently active on this platform

• For more info: Github Link, Video Demo

September 2019

Java, JavaFX, Socket

• A multiplayer chess game built using Java

• This is our term project of Level 1 term 2 at BUET

• You can create a live chess game, join with a awaiting competitor or watch a live chess game as a spectator

• All the matches are saved in memory so that they can be watched later on.

• All legal moves of a chess piece are shown on the board

• The game ends whenever a checkmate occurs

• There can be multiple matches on a single server at any time

• For more information: Github Link, Video Demo

January 2018

ATMega32A, Nodemcu, RFID card

• CSE 316 Microcontrollers and Microprocessors Sessional Final Project 

• We have implemented a card based digital monetary system. 

• Here we have used an ATmega32A micro controller as our processing unit, ESP8266 Nodemcu v2 as wi-fi module for communicating with the remote java server and an RC-522 RFID module for reading from an RFID card. 

• We have also used a 20x4 LCD module for showing prompts and all the inputs are given via a 4x4 keypad.

• For more information: Github Repository, Video Demo

December 2017

Arduino Uno,  Matlab, servo motors

• We made a fully automated robotic hand that could detect and pick up light-weight boxes and sort them into three different categories according the shape that was printed on the boxes.

• The body is made out of PVC boards

• Joints are made by 30Kg-cm Metal gear servo motors

• A 1080p webcam for capturing live images for box detection

• The actuators are controlled by an Arduino Uno

• The arduino is controlled from Matlab

• Whenever a white box of 10cm X 10cm X 10cm appears in front of the webcam, the arm detects and picks up the box automatically

• The arm was configured to detect three shapes (circle, rectangle and triangle) on a single box and is configured to sort the boxes according to the detected shape

• For more information: Github Repo, Video Link