Dmytro Mishkin
Computer Vision at HOVER Inc.
Czechia
Summary
🤩
Rockstar🎓
Top SchoolDmytro Mishkin is a computer vision researcher and engineer with 12 years of experience blending deep learning, classical vision and geometric methods to build real-world systems rather than just benchmark wins. Currently a Senior CV Engineer at HOVER and a postdoc at CTU Prague, he has led research that produced widely-used tools like HardNet and AffNet and contributed performance and feature enhancements to foundational open-source projects such as Caffe and Kornia. He has founded and coordinated an academic research group supporting Ukrainian science, and consulted with startups to align CV research with business and sustainability goals. Comfortable moving ideas from papers to production, he focuses on robust, geometry-aware solutions (e.g., wide-baseline matching under extreme appearance changes) and on managing prototype-engineering teams. His background spans academia and industry, including internships at Intel Labs and a CTO role building mobile visual commerce systems with a pending patent.
12 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD, Artificial Intelligence, Doctor of Philosophy - PhD, Artificial Intelligence at Czech Technical University in PragueComputer Vision, Computer Vision at Czech Technical University in Prague, Faculty of Electrical EngineeringAerospace, Aeronautical and Astronautical Engineering, Aerospace, Aeronautical and Astronautical Engineering at Technische Universität BerlinIgor Sikorsky Kyiv Polytechnic Institute
English, Ukrainian, Russian, Czech
Stackoverflow
2,799reputation
221kreached
91answers
1question
image-processing
top-5%
computer-vision
top-5%
opencv
top-5%
Github Skills (23)
pytorch10
caffe10
geometry10
batch-normalization10
python10
image-processing10
machine-learning10
geometric10
deep-learning10
transform10
computer-vision10
opencv9
cprogramming-language8
c-language8
edge-detection6
Programming languages (14)
Github contributions (5)
kornia/kornia
Jun 2019 - Jan 2023
🐍 Geometric Computer Vision Library for Spatial AI
Role in this project:
Back-end Developer
Back-end DeveloperContributions:2 releases, 245 reviews, 395 commits in 3 years 7 months
Contributions summary:Dmytro made several commits focused on enhancing the core functionality of the geometric computer vision library. They implemented and refined the Gaussian blur and perspective transformations, contributing to the efficiency of image processing algorithms. Furthermore, the user introduced and tested the functionality for image matching, indicating an engagement with the library's core operations related to feature detection and descriptor calculations. These modifications show a focus on improving core geometric and computer vision functionalities.
pytorchdifferentiablepythonvisiondeep-learning
BVLC/caffe
Sep 2014 - Jan 2016
Caffe: a fast open framework for deep learning.
Role in this project:ML Engineer
ML EngineerContributions:14 commits, 13 PRs, 133 comments in 1 year 3 months
Contributions summary:Dmytro contributed to the Caffe deep learning framework, focusing on optimization and feature enhancements. Their work involved removing unnecessary computations in the MVN layer, fixing parsing issues in the log analysis script, and adding a batch normalization layer, including test cases and examples. Furthermore, the user added a channelwise affine layer improving the framework's functionality.
pytorchvisiondeep-learningmachine-learningcaffe
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.