-
learning has strong potential for computer vision, from hyperbolic image segmentation [2] to hyperbolic tree embeddings [3] and hyperbolic vision-language models [4,5]. [1] Nickel, Maximillian, and Douwe
-
geometry altogether and operate in hyperbolic space. Our lab has published multiple papers showing that hyperbolic deep learning has strong potential for computer vision, from hyperbolic image segmentation
-
for deployment and operation on a real-world quantum network prototype, build by the Quantum Internet Alliance (QIA) collaboration. The successful candidate enjoys experimenting with different approaches, and is
-
image description systems tailored for blind and visually impaired (BVI) readers, addressing the diverse and evolving needs of users with different types and levels of visual and cognitive diversities
-
to simulate cascading disaster effects, as well as satellite and sensor data, looking specifically at 6 different use cases across Europe. In addition to the detailed innovative analysis of existing methods and
-
, as well as satellite and sensor data, looking specifically at 6 different use cases across Europe. In addition to the detailed innovative analysis of existing methods and protocols, ITC will focus
-
interactive component. The position is with Prof. dr. Cees Snoek, Professor, head of the Video & Image Sense lab (VIS lab), at the University of Amsterdam. VIS lab is a world-leading lab on Computer Vision and
-
& Image Sense lab (VIS lab), at the University of Amsterdam. VIS lab is a world-leading lab on Computer Vision and Machine Learning, and has over 30 PhD students, postdoctoral researchers and faculty
-
: More and more organizations are diversifying their employee base to stimulate innovation. Diverse groups have members with different, unique knowledge, and by combining and integrating these different
-
infrastructure managers in developing cross-sectoral strategies to proactively shape infrastructure demand, as an alternative to the traditional and increasingly untenable 'predict-and-provide' paradigm. By