• No results found

University of Groningen The snowball principle for handwritten word-image retrieval van Oosten, Jean-Paul

N/A
N/A
Protected

Academic year: 2021

Share "University of Groningen The snowball principle for handwritten word-image retrieval van Oosten, Jean-Paul"

Copied!
3
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

University of Groningen

The snowball principle for handwritten word-image retrieval

van Oosten, Jean-Paul

DOI:

10.33612/diss.160750597

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2021

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

van Oosten, J-P. (2021). The snowball principle for handwritten word-image retrieval: The importance of labelled data and humans in the loop. University of Groningen. https://doi.org/10.33612/diss.160750597

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

A C K N O W L E D G E M E N T S

When I started my PhD research, almost 10 years ago, I didn’t really know what to expect. I know I didn’t expect it to take so long, or that I’d finish a good portion of the work while working a full-time job (for anyone that’s reading this, and considering to do a part-time PhD while working: please reconsider!). However, there were quite a few marvellous people along the way that helped me on this interesting journey.

Of course, I want to thank my supervisor and promoter Lambert Schomaker. I’m very grateful for the opportunity to continue the research that we started with the Master’s project. The many hours that we’ve spent always seemed filled with interesting thoughts and even philosophical debates. It seemed that the more we got into the subject, the longer and deeper our discus-sions became. I’d also like to extend my thanks to the reading committee, Antal van den Bosch, Michael Biehl and Rolf Ingold. I’m also deeply thankful for the opportunity to travel to Japan and visit the Kyushu university. For that, I cannot thank Seiichi Uchida and Volkmar Frinken enough, for the invitation and the interesting scientific discussions, the amazing trips we took and the friendship. This is an experience that will remain a fond memory for a long time.

There are also the many colleagues at the ALICE institute, some of which I shared an office with. It is probably impossible to name everyone, but in particular, I’d like to thank Amir, Aswin, Bea, Ben, Charlotte, Faik, Harmen, Hedde, Jolie, Jort, Mahya, Michiel, Olarik, Robert, Sheng, Tapan, Tijn, and Trudy. A special thanks goes out to Elina for all her excellent help all these years! Then there are the many colleagues at Target Holding / Slimmer AI. I will not name everyone, but I’d be remiss if I didn’t thank

(3)

124 a c k n o w l e d g e m e n t s

Gert-Jan, for all the support and occasionally the kick in the butt, and JC for taking over the supporting role and kicks-in-the-butt these last two years.

Over the course of 10 years, there were a lot of friends that took an interest in my progress and are hopefully also excited for me to finally finish. Some people I’d like to thank in random order: Frank (for your programming wisdom, the great collaboration on Flexc++ and generally being a good friend), Robert and Marieke (all the support, the board games, being there all these years!), Richard and Marloes (also lots of board games, the support and the many hours on the phone while we’re physically distant), Bram and Carine (all the discussions deep into the night, all the support and love), Michiel and Kyuhee (for the hours of coding together), Ben and Charlotte (also lots of board games). My friends from high school and their loved ones: Maurits and Swi Yin, Maarten and Celine (hopefully we’ll be able to get back to escaping rooms soon!). Jilles kept not only my mind sane (with amazing movie nights!) but also my body (I really hope I can get back to kicking your ass!). At Target / Slimmer, I also met some amazing friends: Harma (for all the support and long talks about both our work and life), Marc, Mark and Tom (Passie = Bassie!) and Rolf (for your wisdom, grey hairs, amazing Linux bash-skills and computer history knowledge).

Thanks to Bente for the amazing drawing on the cover of the thesis.

Of course, I want to thank my amazing parents, Paul and Harda. They have always been there when things were a bit rough. Thanks for always believing in me! Thanks also to Reinier and Daphne (and Mees!), Oma, and everyone else in the family for the support and love!

Last but certainly not least: Mitsai. Without her support, this thesis would not have been finished. I would not only have lost my mind, but probably also gone hungry. The unconditional support and love is amazing, and I am forever in your debt for sticking it out with me through all these highs and lows. Thanks love, we made it!

Referenties

GERELATEERDE DOCUMENTEN

While the role of the initial state probability distribution is known to be of relatively small importance (especially in left-right topologies such as Bakis, since these models

Two methods of retrieval will be compared: 1) direct retrieval: ranking, in a single step, all instances from the test set with the distance of the image to the centroid of the

Labelled data is important—for both scientific research as well as industry—because it drives the feedback loop and therefore al- lows for improvements in both accuracy and the

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright

To conclude, this thesis discusses human involvement in the handwriting process from three different angles: In the design of machine learning methods, design of feature

Dit was bestudeerd door data te genereren met een bekend globaal optimum, en veel verschillende modellen te laten trainen op deze gegenereerde data.. Zo konden we de afstand tussen

In Frontiers in Handwriting Recognition (ICFHR), 2012 Interna- tional Conference on, pages

In Frontiers in Handwriting Recognition (ICFHR), 2014 14th International Conference on, pages 537–542.. Segmentation-free word spotting with