About

Hi, I'm Sakib Apon, a computer science graduate with a strong interest in Computer Vision & Natural Language Processing. I'm very passionate and dedicated to my work. I have a strong passion for creativity and discipline. I learned a lot from the open-source community and I love how collaboration and knowledge sharing happens through open-source. I take great care in the experience, architecture, and code quality of the things I build. I am determined to ensure 100% quality for work and committed to deadline.

AI Engineer & Frontend Web Developer.

I am passionate to learn more and get my skill sets more polished and learn new technology.

  • Birthday: 15th May 1997
  • Phone: +880 1951445484
  • City: Dhaka, Bangladesh
  • Age: 24
  • Degree: Bachelors
  • Email: sakibapon7@gmail.com
  • Freelance: Available

Skills

Languages

  • Python
  • Java
  • C++

AI

  • Machine Learning
  • Deep Learning
  • Computer Vision
  • NLP

Library

  • Pandas
  • NumPy
  • Tensorflow
  • Keras
  • PyTorch
  • Open3D
  • Scikit-learn
  • Matplotlib
  • Seaborn

Web Programming

  • HTML
  • CSS
  • Javascript
  • PHP
  • MYSQL

Web Programming Frameworks

  • Bootstraps
  • React
  • Node.js

Education & Experience

Professional Experience

Machine Learning Engineer

Bista Solutions Inc.

  • Working on Projects related to ML & DL.

Machine Learning Engineer

Polyfins Technology Ltd

  • Working on Projects related to ML & DL.

Software Engineer(AI)

Hawar IT

  • Working on Projects related to Computer Vision & GIS.

Research Assistant

Animo AI

  • Working on projects related to Computer Vision.

ML Engineer

Omdena

  • Working on EmpowerYu Project.

AI Engineer

Opus Technology Limited

Dhaka, Bangladesh.

  • Working on live project related to computer vision.
  • Building & Deploying AI models from scratch.

FrontEnd Developer

ISSM

Dhaka, Bangladesh.

  • Developing web application from scratch.

Mentor

BRAC Univeristy

Dhaka, Bangladesh.

  • Served as a mentor in BRAC University English Speaking Activity.
  • Trained freshers to speak fluently in english.

Education

Bsc in Computer Science

2016 - 2021

BRAC Univeristy

Dhaka, Bangladesh.

HSC

2013 - 2015

Jessore MM College

Jessore, Bangladesh.

SSC

2012 - 2013

Jessore Zilla School

Jessore, Bangladesh.

Projects

  • All
  • AI & ML
  • Model & Simulation
  • Web Development

Real-TIme Action Recognition from Video footage

  • Can detect physical bullying in real time.
  • Used DNN with five different image classification models to detect actions from video footage.

Paper Available

CS: GO Action Recognition and Data Collection Automation API

  • Used Five Different Transfer Learning and Majority Voting.
  • Self-developed deep neural network model have also been used for video analysis.
  • System API allows the automated collection and processing of data which can aid to the later training period and enhance our system's performance.

Paper Available

Demystifying Deep Learning Models for Retinal OCT Disease Classification using Explainable AI

  • CNN model is proposed to identify optical coherence tomography diseases in four classes.
  • In terms of model size, the suggested approach is significantly smaller, appropriate for the usage as a web app to generate real-time OCT diagnostic classification.
  • The model is efficient in terms of memory resource.
  • Using LIME & Grad-CAM we examine the interpretability of the proposed model

Paper Available

Yolo V4 Custom Object Detection With Custom Functions API

  • Used Darknet, Deepsort, Tensorflow & Flask.
  • Can Detect, Track, Count, Print Info, Apply Tesseract OCR.
  • Crop Detections and Save as New Image.
  • License Plate Recognition Using Tesseract OCR.
  • Works on Both Videos & Images.

Self Driving Siumulation

  • Used Udacity self driving car simulator.
  • Used Image Processing with CNN

Advanced House Pricing

  • Used different machine learning algorithms to detect house pricing with more than 80 features.

Geonosis Study Abroad

  • A web application where user can save universities, register for consultations and check their saved universities along with previous consultations, assigned consultant, and consultation status.
  • A functional admin panel with restricted access(only admin can access) section where admin can
    • 1. Add University
    • 2. Remove University
    • 3. Modify University
    • 4. Change consultation status
    • 5. Assign consultant

Advanced Car Parts

  • A web application where user can request services and check their previous request history with status(pending, ongoing, or done)
  • A functional admin panel with restricted access(only admin can access) section where admin can
    • 1.Add Services
    • 2. Delete any existing Services
    • 3. Update the state of customers order.

Research Papers

Realtime Action Recogniction

Authors: Tasnim Sakib Apon, Mushfiqul Islam Chowdhury, Md Zubair Reza, Arpita Datta, Syeda Tanjina Hasan, Md. Golam Rabiul Alam

Abstract : Crime rate is increasing proportionally with the increasing rate of the population. The most prominent approach was to introduce Closed-Circuit Television (CCTV) camera-based surveillance to tackle the issue. Video surveillance cameras have added a new dimension to detect crime. Several research works on autonomous security camera surveillance are currently ongoing, where the fundamental goal is to discover violent activity from video feeds. From the technical viewpoint, this i s a challenging problem because analyzing a set of frames, i.e., videos in temporal dimension to detect violence might need careful machine learning model training to reduce false results. This research focused on this problem by integrating state-of-the-art Deep Learning methods to ensure a robust pipeline for autonomous surveillance for detecting violent activities, e.g., kicking, punching, and slapping. Initially, we designed a dataset of this specific interest, which were 600 videos (200 for each action). Later, we have utilized existing pre-trained model architectures to extract features, followed by classification and accuracy analysis.Also, We have classified our models' accuracy, and confusion matrix on different pre-trained architectures like VGG16, InceptionV3, ResNet50, Xception and MobileNet V2. Among the pre-trained models VGG16 and MobileNet V2 performed better.

Accepted @ International conference on sustainable technologies for industry 4.0

View Paper - IEEE Xplore

CS: GO Action Recognition and Data Collection Automation Using Transfer Learning and Majority Voting

Authors: Tasnim Sakib Apon, Abrar Islam, Md. Golam Rabiul Alam

Abstract : Presently online video games have become a progressively favorite source of recreation and Counter Strike: Global Offensive (CS: GO) is one of the top-listed online first-person shooting games. Numerous competitive games are arranged every year by Esports. Nonetheless, (i) No study has been conducted on video analysis and action recognition of CS: GO game-play which can play a substantial role in the gaming industry for prediction model (ii) No work has been done on the real-time application on the actions and results of a CS: GO match (iii) Game data of a match is usually available in the HLTV as a CSV formatted file however it does not have open access and HLTV tends to prevent users from taking data. This manuscript aims to develop a model for accurate prediction of 4 different actions and compare the performance among the five different transfer learning models with our self-developed deep neural network and identify the best-fitted model and also including major voting later on, which is qualified to provide real time prediction and the result of this model aids to the construction of the automated system of gathering and processing more data alongside solving the issue of collecting data from HLTV.

Accepted @ 13th International Conference on Information & Communication Technology and System

View Paper - IEEE Xplore

Demystifying Deep Learning Models for Retinal OCT Disease Classification using Explainable AI

Authors: Tasnim Sakib Apon, Mohammad Mahmudul Hasan, Abrar Islam, Md. Golam Rabiul Alam

Abstract : In the world of medical diagnostics, the adoption of various deep learning techniques is quite common as well as effective, and its statement is equally true when it comes to implementing it into the retina Optical Coherence Tomography (OCT) sector, but (i)These techniques have the black box characteristics that prevent the medical professionals to completely trust the results generated from them (ii)Lack of precision of these methods restricts their implementation in clinical and complex cases (iii)The existing works and models on the OCT classification are substantially large and complicated and they require a considerable amount of memory and computational power, reducing the quality of classifiers in real-time applications. To meet these problems, in this paper a self-developed CNN model has been proposed which is comparatively smaller and simpler along with the use of Lime that introduces Explainable AI to the study and helps to increase the interpretability of the model. This addition will be an asset to the medical experts for getting major and detailed information and will help them in making final decisions and will also reduce the opacity and vulnerability of the conventional deep learning models.

Accepted @ 8th IEEE Asia Pacific Conference on Computer Science and Data Engineering

View Paper - IEEE Xplore

Certificates

<

Contact

Location:

Dhaka, Bangladesh

Call:

+880 1951445484

Loading
Your message has been sent. Thank you!