LAESI: Leaf Area Estimation with Synthetic Imagery
Cited by 4, 2024 Jacek Kałużny, Yannik Schreckenberg, Karol Cyganik, Peter Annighöfer, Sören Pirk, Dominik L. Michels, Mikolaj Cieslak, Farhah …
View Project →Software Development
Cited by 4, 2024 Jacek Kałużny, Yannik Schreckenberg, Karol Cyganik, Peter Annighöfer, Sören Pirk, Dominik L. Michels, Mikolaj Cieslak, Farhah …
View Project →
Cited by 46, 2021 Bosheng Li, Jacek Kałużny, Jonathan Klein, Dominik L. Michels, Wojtek Pałubicki, Bedrich Benes, Sören Pirk We …
View Project →I created IzRise, an AI iPhone app for image resizing and denoising, enabling users to transform low-quality images into high-resolution …
View Project →
I independently designed and developed the website of bookstore IW Jerozolima, creating an online platform to showcase and sell a …
View Project →
I designed and built a mobile app for a Catholic youth event in Poland, providing participants with event details, schedules, …
View Project →
I created the website for "Wspólnota Odnowy w Duchu Świętym Jerozolima", providing an online space for the community's activities and …
View Project →
To properly test the Petifit app during development, I created a fictional dataset of 15 detailed pet profiles and their medical records. This allows me to repeatedly test adding, removing, and displaying pets — without real user data. These characters (like Crookshanks and Garfield) make testing faster, safer, and a lot more fun.
Petifit is designed to help pet owners manage medical records, vaccines, medications, chip numbers, and more. But building this kind of app means lots of testing, and I needed sample data that was repeatable, safe, and realistic — but not real.
Using fictional pets from well-known books and media, I built a rich dataset of 15 animals. They each have complete profiles, fake chip numbers, realistic veterinary stamps, and quirks designed to test edge cases — from extreme weights to multilingual names and weird formatting.
No real pet data, chip numbers, or veterinary information at risk
Consistent test data that I can rely on across multiple development cycles
Designed to test unusual scenarios, long names, and formatting edge cases
Perfect for edge cases: Hermione Granger's famously half-Kneazle cat tests non-standard breeds, magical backgrounds, and unusual physical traits.
His health book includes multilingual notes and long-winded remarks, perfect for testing font scaling, OCR behavior, and overflow cases in both English and Polish.
Stress testing champion: His exaggerated weight (12kg), fictional chip, and sarcastic medical notes like "lasagna overdose" make him ideal for testing field limits.
Garfield's file is heavy on long text and silly treatments, perfect for deletion and re-import tests. If something breaks during development, it's probably Garfield's fault!
This dataset also prepares for the next phase: training AI to read these health books automatically. Petifit will eventually scan these images and turn them into structured health records using advanced OCR technology.
Access the full collection of 15 fictional pets plus 15 health book scans
Feel free to add your own creations! Let's make it better!
📁 Dataset