Hovnatan Karapetyan
Verified Expert in Engineering
Computer Vision Developer
Hovnatan is a computer vision engineer with a PhD in applied physics from Stanford. He is passionate about computer vision and image processing. Hovnatan has worked on a comprehensive variety of projects in the field. He is also a well-rounded software engineer, having integrated his solutions both in cloud-based and desktop (Windows, MacOS) environments.
Portfolio
Experience
Availability
Preferred Environment
Visual Studio Code (VS Code), Linux, MacOS, PyTorch, OpenCV, Vim Text Editor
The most amazing...
...thing I’ve built used computer vision and image processing to detect weeds and other problematic areas of agricultural fields using aerial imagery.
Work Experience
Computer Vision Engineer
Stealth Technology Startup
- Developed a semantic segmentation deep learning model.
- Oversaw data generation and annotation for a semantic segmentation deep learning model.
- Integrated deep learning models into Electron, Unity, and desktop native projects using Open Neural Network Exchange (ONNX) and Core ML technologies.
- Researched video action recognition methods and models.
Computer Vision Engineer
SuperAnnotate
- Developed a video object segmentation deep learning algorithm with PyTorch.
- Integrated deep learning model inference with Amazon SageMaker.
- Led development of a Python SDK to the company platform.
Computer Vision Engineer
IntelinAir
- Developed a weed detection algorithm with high resolution aerial images of agricultural fields in US.
- Implemented performance critical memory optimized and multithreaded C++ image processing code that worked on huge 10,000x10,000 images.
- Improved image registration code using multithreading.
Software Engineer
Mentor Graphics (now Siemens)
- Developed various components in an electronic design automation (EDA) internal database, using C++.
- Fixed numerous bugs and maintained legacy C++ code.
- Refactored to better software engineering standards and tools, using Boost.
Data Scientist
PicsArt
- Developed a collaborative filtering based photo recommendation system for the PicsArt photo sharing social network.
- Analyzed the usage of the PicsArt mobile app's various components using HBase and MapReduce.
- Analyzed user behavior on the social network using SQL and Pandas.
Experience
Weed Detection Using Aerial Imagery
https://www.intelinair.com/Skills
Languages
Python, C++, SQL, Java, Scala, Objective-C, JavaScript, C#
Libraries/APIs
PyTorch, OpenCV, NumPy, GEOS, Pandas
Tools
Vim Text Editor, MATLAB, Amazon SageMaker
Platforms
Visual Studio Code (VS Code), Linux, MacOS, Amazon Web Services (AWS)
Other
Machine Learning, Computer Vision, Applied Mathematics, Deep Learning, Image Processing, Artificial Intelligence (AI), Satellite Images, Aerial Photography, Fourier Analysis, Object Detection, Semantic Segmentation, Image Registration, Electronic Design Automation (EDA), Recommendation Systems, Open Neural Network Exchange (ONNX), Applied Physics, SDKs, Object Tracking, Object Segmentation
Frameworks
Hadoop, Spark, Unity, Core ML, Boost
Paradigms
MapReduce
Storage
HBase
Education
PhD Degree in Physics
Stanford University - Stanford, CA
Bachelor's Degree in Applied Mathematics and Physics
Moscow Institute of Physics and Technology - Moscow, Russia
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring