|June 2018:||State-of-the-art accuracy in automatic plant and fungi identification confirmed by scoring 1st in the new Fine-Grained Visual Categorization challenges: FGVCx Flower Classification and FGVCx Fungi Classification. The results were be presented at the FGVC5 workshop at CVPR 2018.|
|May 2018:||Our plant recognition system scored 1st in the ExpertLifeCLEF 2018 plant identification challenge, achieving 88.4 % accuracy in recognition of 10,000 plant species! Tested against human experts in plant sciences, our system scored better than half (5/9) of the experts.|
|December 2017:||Our article on Fine-grained Recognition of Plants from Images was published in the Plant Methods' special issue on Plants in Computer Vision.|
|October 2017:||Our work was presented in the keynote of Jiří Matas at the CVPPP workshop at ICCV 2017.|
|September 2017:||I presented our contribution to LifeCLEF 2017, which scored 3rd in the plant identification task, at the Conference and Labs of the Evaluation Forum, 2017.|
|May 2017:||I passed my PhD state exam and I started my internship at Google, Los Angeles, in the Mobile Vision team.|
|January 2017:||I visited The 1st Winter School in Computer Science and Engineering on Computer Vision in Jerusalem and presented our research.|
|November 2016:||I had a talk and demo at the BMVA technical meeting on Plants in Computer Vision in London, 15. November 2016. [Blog post from Hannah Dee].|
|October 2016:||I attended to the Google Computer Vision PhD Summit in Zurich, 17.-19. October 2016.|
|September 2016:||Our US patent no. 9,443,164, "System and Method for Product Identification", was officially issued by USPTO. Presentation of our 3rd best results in the LifeCLEF 2016 image-based plant identification task at the Conference and Labs of the Evaluation Forum, 2016.|
M. Sulc and J. Matas, Learning with Noisy and Trusted Labels for Fine-Grained Plant Recognition.
In Working Notes of CLEF 2017 - Conference and Labs of the Evaluation Forum, 2017.
M. Sulc, D. Mishkin and J. Matas, Very Deep Residual Networks with MaxOut for Plant Identification in the Wild.
In Working Notes of CLEF 2016 - Conference and Labs of the Evaluation Forum, 2016.
M. Sulc and J. Matas, Significance of Colors in Texture Datasets.
In Proceedings of the 21st Computer Vision Winter Workshop, 2016.
M. Sulc, A. Gordo, D. Larlus and F. Perronnin, System and Method for Product Identification. [Link]
US Patent No. 9,443,164.
Issued in August 2016.
M. Sulc and J. Matas, Fast Features Invariant to Rotation and Scale of Texture.
[Springer Link] [PDF]
European Conference on Computer Vision (ECCV) 2014 Workshops (LBP'14). Springer International Publishing, 2014.
M. Sulc and J. Matas, Texture-Based Leaf Identification.
[Springer Link] [PDF]
European Conference on Computer Vision (ECCV) 2014 Workshops (CVPPP'14). Springer International Publishing, 2014.
M. Sulc and J. Matas, Kernel-mapped Histograms of Multi-scale LBPs for Tree Bark Recognition.
In Proceedings of the 28th Conference on Image and Vision Computing New Zealand, 2013.
The course materials for the next labs are available on the corresponding CourseWare page (Czech).
Prize of Josef Hlávka for the Best Students and Graduates
The Foundation of Josef, Marie and Zdeňka Hlávka, 2014
Best Internship Project Presentation
Xerox Research Centre Europe, 2014
Dean's Prize for Outstanding Master's Thesis
FEE CTU in Prague, 2014
Prize of the Masaryk Institute Director for an Original and Precise Thesis.
MIAS CTU in Prague, 2015