no code implementations • 21 Nov 2023 • Christian Keilstrup Ingwersen, Anders Bjorholm Dahl, Janus Nørtoft Jensen, Morten Rieger Hannemose
We propose a novel loss function, multiview consistency, to enable adding additional training data with only 2D supervision.
1 code implementation • 4 Apr 2023 • Christian Keilstrup Ingwersen, Christian Mikkelstrup, Janus Nørtoft Jensen, Morten Rieger Hannemose, Anders Bjorholm Dahl
Accurate 3D human pose estimation is essential for sports analytics, coaching, and injury prevention.
no code implementations • 4 Apr 2023 • Patrick Møller Jensen, Vedrana Andersen Dahl, Carsten Gundlach, Rebecca Engberg, Hans Martin Kjer, Anders Bjorholm Dahl
Annotated training data is essential for deep learning-based object detection.
no code implementations • 30 Nov 2020 • Wail Mustafa, Christian Kehl, Ulrik Lund Olsen, Søren Kimmer Schou Gregersen, David Malmgren-Hansen, Jan Kehres, Anders Bjorholm Dahl
Furthermore, the results indicate that the network can exploit the coupling of the channels to enhance the overall quality and robustness.
no code implementations • 28 Oct 2018 • Christian Kehl, Wail Mustafa, Jan Kehres, Anders Bjorholm Dahl, Ulrik Lund Olsen
Multi-spectral computed tomography is an emerging technology for the non-destructive identification of object materials and the study of their physical properties.
1 code implementation • 6 Sep 2018 • Vedrana Andersen Dahl, Monica Jane Emerson, Camilla Himmelstrup Trinderup, Anders Bjorholm Dahl
Since we combine a limited amount of user-labelled data with the clustering information obtained from the unlabelled parts of the image, our method fits in the general framework of semi-supervised learning.