University of West BohemiaFaculty of Applied SciencesCzechia
Office: NTIS, Technická 8, 301 00 PilsenRoom UN525 | Map

Projects

Danish Svampe Atlas

  • Developing Rest API service for automatic fungi recognition.
  • Fine-tuning of the lightweight MobileNet-V3 custom architecture.
  • In close cooperation with Czech Technical University, Danish Mycological Society, Gbif.org and Google.
  • iOS App | Android App | Web App | Models

Snake Species Recognition

  • A platform for fast and reliable snake species recognition using AI and herpetologists..Developing REST API service and Web Application for automatic snake species recognition. Curatingnovel open-source benchmark dataset and organizing ML competitions
  • In close cooperation with Florida Gulf Coast University, University of Geneva, Instutute of Global Health, Médecins Sans Frontières and many more.
  • About Project | Demo App

Varroa Destructor Detection

Lynx ID

  • Research & Development in the topic of Lynx identification.
  • In close cooperation with Mendel University in Brno, Bavarian Forest National Park, Leibniz-Institute for Zoo and Wildlife Research and many others.
  • About Project

Competitions & Challenges

  • 1st place in the ImageCLEFdrawnUI Challenge. [2021]
  • 1st place in the Crop Disease recognition - ICLR Computer Vision for Agriculture Workshop. [2020]
  • 1st place in the Hakuna Ma-Data recognition challenge sponsored by Microsoft AI for Earth [2020]
  • 1st place in the ImageCLEFcoral Challenge. [2020]
  • 1st place in the ImageCLEFdrawnUI Challenge. [2020]
  • 1st place in the FGVCx Flower Classification and FGVCx Fungi Classification. [2019]
  • 2nd place in the FGVCx iDesigner. [2019]

Publications

Fungi Recognition: A Practical Use Case
M. Šulc, L. Picek, J. Matas, T. Jeppesen, J. Heilmann-ClausenIEEE WACV | 2020 | URL
Coral Reef Annotation, Localisation and Pixel-wise Classification using Mask-RCNN and Bag of Tricks
L.Picek, A. Říha, A. ZitaCLEF | 2020 | URL
Sketch2Code: Automatic hand-drawn UI Elements Detection with Faster-RCNN
A. Zita, L.Picek, A. ŘíhaCLEF | 2020 | URL
Mastering Large Scale Multi-label Image Recognition with high efficiency over Camera trap images
M. Valan, L. PicekCVPR - FGVCW | 2020 | URL
Overview of LifeCLEF 2020: a System-oriented Evaluation of Automated Species Identification and Species Distribution Prediction
Book Chapter | 2020 | URL
Overview of the SnakeCLEF 2020: Automatic snake species identification challenge
L.Picek, I. Bolon, A. Durso, R. CastañedaCLEF | 2020 | URL
Recognition of the Amazonian Flora by Inception Networks with Test-time Class Prior Estimation
L. Picek, M. Šulc, J. MatasCLEF | 2019 | URL
Plant Recognition by Inception Networks with Test-time Class Prior Estimation
M. Šulc, L. Picek, J. MatasCLEF | 2018 | URL

Media

  • [ENG] - Meet the winners of the HaKuna Ma-Data challenge - URL
  • [ENG] - Doctors Without Borders to field test Czech researcher’s snakebite app - URL
  • [ENG] - Google AI Blog Announcing the 7th Fine-Grained Visual Categorization Workshop - URL
  • [CZ] - Čeští vědci vyvinuli software pro mobilní aplikaci, která rozeznává houby - URL
  • [CZ] - Student ZČU v Plzni Lukáš Picek sklízí mezinárodní úspěchy - URL
  • [CZ] - Česká umělá inteligence pozná houby stejně dobře jako nejlepší experti - URL