Ali Ghofrani
Deep Learning Researcher
Python Developer
Lavizan, Tehran, Iran
(-
Personal website: http://ghofrani.ir/
Email :-,-Linkedin : https://www.linkedin.com/in/ali-ghofrani/
GitHub Repo : https://github.com/alighofrani95
Google Scholar: https://scholar.google.com/citations?user=yGPdxcYAAAAJ&hl=en&oi=sra
Stackoverflow: https://stackoverflow.com/users/-/ali-ghofrani
INTRODUCTION
Experienced Machine Vision and Deep Learning researcher with a demonstrated history of
working in the Image Processing and Artificial Intelligence industries.
Skilled in Python, OpenCV, Sci-Kit Learn, TensorFlow, HuggingFace, NLTK, FastAPI, MongoDB,
PostgreSQL, and TFlite. Strong research professional with an MSc focused in Signal Processing
and Communications Engineering.
As an ML Developer, I'm interested in training from scratch or finetuning state-of-the-art deep
architectures based on TensorFlow, PyTorch, or MXnet for different Vision, Speech, or NLP tasks.
But due to my 5 years of experience of real-world challenges, As an MLOPs Developer, I'm in love with
Quantize, Prune, and Distill my trained model to lighter and faster production-ready microservices
based on TVM, TFlite, ONNX, OpenVino, and CoreML to bring huge models to server-side or client-side
applications.
PUBLICATIONS
CaTILoc: Camera Image Transformer for Indoor Localization [link]
The International Conference on Acoustics, Speech, & Signal Processing (ICASSP 2021)
APS: A Large-Scale Multi-Modal Indoor Camera Positioning System (Conference Best Paper Award) [link]
The 4th Mediterranean Conference on Pattern Recognition and Artificial Intelligence (MedPRAI 2020)
Plant Disease Recognition using Optimized Deep Convolutional Neural Networks (Oral Presentation) [link]
The 4th Mediterranean Conference on Pattern Recognition and Artificial Intelligence (MedPRAI 2020)
L-ICPSnet: LiDAR Indoor Camera Positioning System for RGB to Point Cloud Translation using End2End
Generative Network (Oral Presentation) [link]
The 8th Iranian Joint Congress on Fuzzy and Intelligent Systems (CFIS)
Attention-Based Face AntiSpoofing of RGB Camera using a Minimal End-2-End Neural Network
(Oral Presentation) [link]
2020 International Conference on Machine Vision and Image Processing (MVIP)
ICPS-net: An End-to-End RGB-based Indoor Camera Positioning System using deep convolutional neural
networks (Oral Presentation) [link]
The 12th International Conference on Machine Vision, Amsterdam, The Netherlands.
Real-time Face-Detection and Emotion Recognition Using MTCNN and miniShuffleNet V2
(Oral Presentation – Best Paper Candidate) [link]
5th Conference on Knowledge-Based Engineering and Innovation [IEEE KBEI2019]
Capsule-Based Persian/Arabic Robust Handwritten Digit Recognition Using EM Routing
(Oral Presentation) [link]
Sharif 4th International Conference on Pattern Recognition and Image Analysis [IEEE IPRIA2019]
PATENTS
ROBUST CAMERA LOCALIZATION BASED ON A SINGLE COLOR COMPONENT IMAGE AND
MULTI-MODAL LEARNING
International Application No. PCT/IB2020/042451
METHOD AND SYSTEM FOR FACE ANTI-SPOOFING BASED ON DEEP LEARNING,
RESISTANT TO A WIDE VARIETY OF ATTACKS
National Application No-
METHOD FOR CAMERA LOCALIZATION AND SYSTEM THEREOF
National Application No-
Volunteer Academic Service
It’s my honor to be one of the reviewers of the “Journal of Photogrammetry and Remote Sensing” on
Q1 Elsevier journal with an impact factor of 8.97 and also be a reviewer in the “Computational
Imaging” field at the ICASSP 2022 conference.
EXPERIENCE
NSAtech, Doha Qatar - CTO
http://nsatech.co/
Feb 2020 - Sep 2021
●
●
●
Develop action similarity system via 2d Skeleton and 3D body mesh reconstruction via simple RGB
camera as an input to compare amateur and professional players.
Train, Evaluate, and Serve deep learning models to get ready trained models to deploy on servers
Design and develop APIs to interact UI with backend services
** This startup has been selected as one of the top 10 startups during cohort 4 of the QatarSportsTech Accelerator
program among more than 280 startups from 38 countries. [link]
Alpha Reality, Tehran, Iran / Toronto, Canada - VP of AI Dept
http://alphareality.io/
Jan 2019 - PRESENT
●
Deploy Face Recognition and Face AntiSpoofing via Iranian Nation ID Card API (SabteAhval) for
Sejam Stock Market Electronic Authentication (SahmAshena Brokerage)
●
AlphaCognition Project: VIP and BlackList Face Recognition System at MegaMarkets from Multiple
CCTV Streams. (ShahrVand Chain Stores)
●
Retail market analysis engine and recommender system based on customer segmentation and
behavior prediction (Golrang System)
Iran's National Elites Foundation - Deep Learning Developer
Dec 2019 - PRESENT
Persian Sign Language Project:
●
●
●
●
Converting Video Signs to the corresponding text (Gloss) labels
Converting Gloss to the Pure Farsi Texts (Natural Machine Translation Task)
Using the SOTA Language Model such as BERT, XLM, GPT-2, and Attention approaches.
Pre-training from scratch “GPT-neo” and “DistilRoBERTa” language models on more
than 125GB of pre-processed self-crawled Persian Text via Google TPUv3-32 on 12 days
(Thanks to Huggingface and Google Research for hardware resource sponsoring)
These models outperform previous results on downstream Persian NLP tasks such as NER,
Sentiment Classification, Text Classification, and so on.
Alpha Reality, Tehran, Iran / Toronto, Canada - Computer Vision Developer
Apr 2019 - Jan 2020
●
Indoor Navigation via Camera Positioning based on End to End Deep Learning approach using
State of the art EfficientNet and MobileNetV2 CNN architectures. (©navify.app) [link] [link]
●
Implementation Face Recognition with Liveness Detection Based on Deep Learning methods.
●
Face AntiSpoofing WebService based on Face 3D Pose Estimation for Electrical Authentication.
Aradow, Tehran, Iran - Machine Learning Developer
Jun 2018 - Apr 2019
●
Implementation of the diagnostic system for livestock diseases based on the given symptoms
Server-side microservice as docker container using SVM Gaussian kernel and RESTful API
(Sci-Kilt Learn, Flask)
●
Implementation of the plant disease recognition system based on the given client-side leaf image
Using the CNN MobileNetV2 Architecture (Keras, Flask)
●
Implementation of the Recommendation system based on user preferences and their comment
sentiments (Extracting information from MongoDB)
●
●
Prediction of harvesting rates based on land climatic features and past years’ reports.
Crop recommendation system based on nutrient report and soil features
IRIB, Jame Jam - Machine Vision Researcher
Aug 2017 - Jun 2018
●
●
●
Implementation of Real-Time Face Detection and Emotion Recognition based on Deep Learning
Converts low frame rate videos to Super-SloMo via interpolating mid-frames of video
Detecting faces hidden in covers such as face masks, helmets, military equipment, and ...
SafeBana Name, Tehran - Computer Network Admin NOC
Sep 2016 - Aug 2017
●
●
Implementation of a comprehensive monitoring system under the SNMP protocol using PRTG
(among the branches of the Company)
Optimizing network infrastructure and hierarchical security enhancement at the central branch
office (Based-on Cisco Security Checklist)
EDUCATION
Isfahan University of Technology, Isfahan - B.Sc.
Sep 2013
IRIBU, Tehran - B.Sc. Electrical, Electronics and Communications Engineering
Jan 2014 - April 2017
Thesis: Real-Time Face Detection & Emotion Recognition. (Best Thesis Award)
GPA: 18.3/20
IRIBU, Tehran - M.Sc. Telecommunication Systems
Sep 2017
Thesis: Camera reLocalization based on MultiModal LiDAR and RGB Deep Neural Networks
Using the brilliant talent quota for direct admission (Exceptional Talents Development Center)
Licenses & certifications
TensorFlow: Advanced Techniques Specialization
DeepLearning.AI
[Link] CredentialID: 6V6NQ3RSGJLS
Generative Adversarial Networks (GANs) Specialization
DeepLearning.AI
[Link] CredentialID: PNHZZ55BWLJU
Deep Learning with TensorFlow
IBM CognitiveClass.ai [Link] CredentialID: f2ea-e43f5aebf2edce3d42057
Machine Learning with Python
IBM CognitiveClass.ai [Link] CredentialID: 0c5dd990d6014acc974b4e2f5205d26c
AI From the Data Center to the Edge – An Optimized Path Using Intel® Architecture
Intel® Corporation [Link]
Natural Language Processing with Attention Models
Coursera [Link] Credential ID: YK34TWS6LPJR
Challenges & Awards
Won 2nd place and the best algorithm award in the time series traffic forecasting of Neshan Navigation
Application.
Lectures & Presentations
Deep Learning Concepts and Implementation by Keras
MEETS, IRIB University, College Students' Scientific Association
Artificial Neural Networks And Machine Vision [Link]
The Brain & Cognition association of Guilan University, As Invited Lecturer
Python Course, [As Invited Lecturer at IHU ]
SKILLS
AI & ML Field
Having the experience of developing with C, Matlab, and Expert in Python
Having the experience of implementing regression, classification, clustering algorithms by Sci-Kit Learn
framework in a Python environment
Having the experience of implementing artificial neural network and deep learning architectures by Keras
and TensorFlow frameworks (and also familiar with PyTorch, MXnet, GluonCV, Caffe)
Having the experience of using Numpy, Matplotlib, Pandas for preprocessing, normalization, and
visualization of data on big scales
Having the experience of implementing deep learning-based machine vision methods for Object Tracking,
Object Detection, and Object Recognition by OpenCV framework (Familiar with Faster R-CNN, YOLO,
SSD)
Having the experience of Deep Learning based algorithms for video classification based on I3D, SlowFast,
and non-local architectures.
Having the experience of implementing deep learning-based machine vision methods for human body
pose such as PoseNet, OpenPose, and AlphaPose.
Having the experience of dockerizing AI applications as microservices
Skilled in Linux OS Distro and Bash Command Line (Especially Debian-based distro such as ubuntu, mint)
Familiar with implementing RESTful APIs by Flask and Django
Familiar with both SQL and NoSQL databases and fetching information (such as MongoDB, Redis, MySQL)
Familiar with JSON and XML data structures
Familiar with ARM-based embedded systems such as RasPi, etc, and Optimization of algorithms to
enhance performance with parallel processing (Intel TBB, GStreamer, ...)
Familiar with working in an Agile (SCRUM) environment
Familiar with using version control such as git, GitLab, and GitHub
Familiar with teamwork applications such as Trello and Jira
Familiar with NLP new approaches such as Google BERT and GPT networks
Computer Network Field
Cisco: Highly skilled in cisco devices configuration [CCNA cert 2016 NooraNet]
Microsoft: Skilled in implementing Microsoft services such as AD, DNS, DHCP, VPN, etc [MCSE2012 cert]
Virtualization: Skilled in server and desktop virtualization and configuration VMWare vSphere (ESXi,
vCenter, HorizonView) and also Microsoft Hyper-V
Monitoring: Skilled in Network Monitoring via SNMP protocol and implementing PRTG Server
MikroTik: Skilled in implementing services such as vlanning, routing, VPN servers, etc.
COMMENTS
●
A fast learner, creative, good team member, and motivated to explore the state-of-the-art tech
●
Interested in researching and developing in the debates of new algorithms and modern
architecture of deep learning and machine vision in open source environments
Interested in extracting and analyzing information in a large scale volume of big data
Interested in topics related to ethical hacking and security implications
Interested in analyzing social network information via deep learning and machine vision concepts
●
●
●
