Humam Helfawi, Developer in Mersin, Akdeniz, Turkey
Humam is currently unavailable

Humam Helfawi

Bio

Humam is a developer with over eight years of experience. He loves taking on complex challenges and is interested in any project where math is translated into code, and the product being developed drives the future forward. Humam is passionate about a number of things, including modern C++, autonomous driving and piloting, 3D inspection, and heterogeneous computing.

Portfolio

Treo AI
OpenCV, C++, Python, PyTorch, You Only Look Once (YOLO), 3D Pose Estimation...
Six2o Cricket
OpenCV, C++, Python, NVIDIA CUDA, PyTorch, NVIDIA TensorRT...
Freelancer
OpenCV, PyTorch, You Only Look Once (YOLO), Maps, Tracking, C++, Python...

Experience

  • Optimization - 5 years
  • C++ - 5 years
  • OpenCV - 5 years
  • 3D - 3 years
  • ESP32 - 2 years
  • Simultaneous Localization & Mapping (SLAM) - 2 years
  • Drones - 1 year
  • Robotics - 1 year

Preferred Environment

Visual Studio, Linux, Windows, Python, Artificial Intelligence (AI), Machine Learning, Deep Learning, Convolutional Neural Networks (CNNs), Image Processing

The most amazing...

...thing I've developed was a single-shot 3D structured light scanner.

Work Experience

Principal Computer Vision Engineer

2024 - PRESENT
Treo AI
  • Developed a real-time golf analysis app for golf courses.
  • Enabled real-time golf shot detection using a custom-made AI model.
  • Built real-time, ball-accurate trajectory tracking from a single static camera.
  • Optimized models and algorithms using CUDA and TensorRT.
  • Transformed pose model distillation into YOLO pose.
Technologies: OpenCV, C++, Python, PyTorch, You Only Look Once (YOLO), 3D Pose Estimation, Pose Estimation, Deep Learning, Tracking, ONNX Runtime, NVIDIA TensorRT, NVIDIA CUDA, Multiprocessing, Image Analysis, Image Segmentation, Video Processing

Senior Computer Vision Engineer

2024 - PRESENT
Six2o Cricket
  • Developed CV-based cricket game analysis and trajectory tracking.
  • Enabled real-time high-performance detection and tracking using TensorRT.
  • Used multiple ZedX cameras for 3D tracking simultaneously.
  • Deployed the project on an edge device (NVIDIA AGX).
Technologies: OpenCV, C++, Python, NVIDIA CUDA, PyTorch, NVIDIA TensorRT, You Only Look Once (YOLO), Deep Learning, Calibration, Tracking, Multiprocessing, ONNX Runtime, NVIDIA Jetson, NVIDIA Jetson AGX Orin, Image Analysis, Video Processing

Senior Computer Vision Engineer

2020 - PRESENT
Freelancer
  • Contributed to AI-based map track compressor and route optimization.
  • Handled real-time document detection and tracking (iOS and cloud).
  • Contributed to multiple-shot dental scan point-cloud registration.
  • Developed skin lesion classification using the deep learning crowd counting model.
  • Implemented hand-drawn sketch detection, segmentation, and classification.
Technologies: OpenCV, PyTorch, You Only Look Once (YOLO), Maps, Tracking, C++, Python, Deep Learning, Image Analysis, Image Segmentation, Video Processing

Computer Vision Expert

2020 - 2023
Strong Analytics, LLC
  • Developed a novel solution for horse localization from a PTZ camera.
  • Enabled depth retrieval and 3D horse positioning using a 3D camera.
  • Implemented panorama stitching of a horse race track using six cameras in near real-time.
  • Contributed to markerless automatic PTZ camera calibration in real time.
  • Enabled automatic simulation of cameras and lens attributes to cover a horse race track.
  • Contributed to a simulation of an RFID system to track horses on the track in real time.
Technologies: Computer Vision, Simultaneous Localization & Mapping (SLAM), Tracking, 3D Vision, Stereoscopic Video, 3D Cameras, Video Stitching, OpenCV, Image Analysis, Video Processing

Research Engineer

2017 - 2020
Optecks, LLC
  • Led the research and development of an autonomous drone SDK called Ottofly.
  • Managed and built a single-shot, 3D, structured light scanner.
  • Constructed an IoT application using ESP32, 8266, and Alexa.
  • Designed novel solutions for robotics localization problems.
Technologies: ArduPilot, MAVLink, Drones, Computer Vision, 3D, C++, OpenCV, Image Analysis, Image Segmentation

Research Engineer

2015 - 2017
IDscan Biometrics, Ltd.
  • Created real-time document detection and tracking for smartphones.
  • Extracted and authenticated IDs and passports' hologram patterns: visible, near-infrared, and ultraviolet IDs authentication.
  • Made online image reconstruction from streaming feeds.
  • Built real-time image registration for documents, which is used by multiple other modules.
  • Enabled document text detection in complex backgrounds.
  • Built C++ generic libraries for the proposed research.
  • Created C# wrappers for C++ core libraries to provide an interface for the other modules.
  • Wrote both detailed and abstract reports for research outcomes.
  • Benchmarked state-of-the-art algorithms on specific platforms.
Technologies: C++, OpenCV, Computer Vision, Image Processing, Image Analysis, Image Segmentation, Video Processing

Experience

Training ANFIS Using the Enhanced Bees Algorithm and Least Squares Estimation

An academic publication in the field of machine learning. Trained ANFIS using the Enhanced Bees Algorithm and least squares estimation. We tried new techniques to train a neuro-fuzzy system and proved its superiority among the other methods.

Education

2009 - 2014

Bachelor's Degree in Informatics Engineering

University of Aleppo - Aleppo, Syria

Skills

Libraries/APIs

OpenCV, PyTorch, Node.js

Tools

You Only Look Once (YOLO), GitHub, MATLAB, Visual Studio, Microsoft Access, TortoiseGit, TFS, GitLab, NVIDIA Jetson

Languages

C++, Visual Basic .NET (VB.NET), C, Python 3, Python, Visual Basic 6 (VB6), C#, Java, Visual Basic for Applications (VBA), XML, CSS, JavaScript, VBScript, HTML, Delphi, Prolog

Frameworks

Boost, ASP.NET, Windows Presentation Foundation (WPF)

Paradigms

Synthetic Data Generation, Agile

Platforms

Windows, Linux, Java RTS, AWS IoT, Amazon Alexa, Amazon Web Services (AWS), NVIDIA CUDA, NVIDIA Jetson AGX Orin

Storage

SQL Server 2008

Other

Computer Vision, Robust Estimation, Simultaneous Localization & Mapping (SLAM), 3D, Stereoscopic Video, Clustering, Neural Networks, Neuro-Fuzzy, Object Detection, Fuzzy Logic, Image Registration, IoT Protocols, Artificial Intelligence (AI), Machine Learning, Image Processing, Image Analysis, Video Processing, Optimization, Robotics, Drones, Mathematics, Text Detection, Image Quality Assessment, Deep Learning, Convolutional Neural Networks (CNNs), Image Segmentation, MAVLink, ArduPilot, ESP32, ESP8266, SVMs, Microsoft 365, Tracking, 3D Vision, 3D Cameras, Video Stitching, Maps, NVIDIA TensorRT, Calibration, Multiprocessing, ONNX Runtime, 3D Pose Estimation, Pose Estimation

Collaboration That Works

How to Work with Toptal

Toptal matches you directly with global industry experts from our network in hours—not weeks or months.

1

Share your needs

Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2

Choose your talent

Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3

Start your risk-free talent trial

Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.

Top talent is in high demand.

Start hiring