Matěj Šmíd
Verified Expert in Engineering
Research Developer
Matěj is a computer vision researcher with experience spanning both academia and industry. After spending five years as an industrial researcher abroad, he returned to earn a Ph.D. in artificial intelligence in a world-class international team at the Czech Technical University in Prague.
Portfolio
Experience
Availability
Preferred Environment
Git, Jupyter, PyCharm, Linux, Research
The most amazing...
...thing I've developed is an extremely precise method to synchronize multiple cameras by exploiting properties of a rolling shutter sensor.
Work Experience
Computer Vision Engineer
Toptal
- Created a tool for road traffic researchers to track vehicles in drone footage.
- Evaluated aggressiveness and cooperativeness of vehicle drivers on a highway merging lane using video analysis.
- Fine-tuned a high-resolution object detector on the customized dataset.
Research Assistant
Czech Technical University
- Researched new methods and applications for multi-object tracking in videos. The methods were applied to track ice hockey players, floorball players, and small laboratory animals.
- Calibrated and synchronized a complicated multi-camera system with zoom and fisheye cameras.
- Redesigned and taught machine learning course labs.
- Developed an audio-based insect counting software for biodiversity monitoring.
Assistant Lecturer
Upper Austria University of Applied Sciences
- Designed labs for a computer vision course from scratch.
- Tutored a computer vision course with a focus on image processing.
- Graded MATLAB exercises and a final project done by the students.
Industrial Researcher
Software Competence Center Hagenberg, GmbH
- Participated in a variety of computer vision projects for companies and industries.
- Contributed to an EU research project with biologists and bioinformatics.
- Created an automated measurement system for objects in volumetric data.
- Developed software and vision hardware for a vision-based measurement device.
Experience
Vision-based Desktop Application for Tracking Laboratory Animals
I took over a large and messy codebase, refactored the majority of the code to be maintainable and well structured, and contributed new features. The application uses deep learning to locate, track, and re-identify multiple individuals in videos.
Multiple Camera Synchronization Method
https://bit.ly/2QHfudgSoftware for Measurements in Optical Coherence Tomography Volumetric Data
Cancer Cell Tracking Project
Floorball Player Tracking
https://www.youtube.com/watch?v=gelNVHccl1MThe method demonstrated in the video uses probabilistic occupancy maps algorithm to fuse data from multiple views and K-shortest paths graph algorithm to find trajectories in the 2D-occupancy maps.
Skillset
Languages
Python, Markdown, C++, C, HTML
Libraries/APIs
Matplotlib, NumPy, OpenCV, Pandas, FFmpeg, ImageMagick, SciPy, PyTorch, Keras
Platforms
Linux, Docker
Other
Object Tracking, Computer Vision, Image Processing, Artificial Intelligence (AI), Image Recognition, Machine Learning, Research, Shell Scripting, Scientific Computing, Object Detection, Calibration, Classification Algorithms, Cython, Deep Learning, Mathematics, Statistics, Linear Regression, 3D
Tools
MATLAB, Git, Jupyter, PyCharm, CMake, Visual Studio, Google Compute Engine (GCE)
Frameworks
Qt
Storage
Google Cloud
Education
Ph.D. Degree (Ongoing) in Artificial Intelligence (Computer Vision)
Czech Technical University - Prague, Czech Republic
Master's Degree in Technical Cybernetics
Czech Technical University - Prague, Czech Republic
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