2 code implementations • 23 Mar 2020 • Vladimiros Sterzentsenko, Alexandros Doumanoglou, Spyridon Thermos, Nikolaos Zioulis, Dimitrios Zarpalas, Petros Daras
This is accomplished by a soft, differentiable procrustes analysis that regularizes the segmentation and achieves higher extrinsic calibration performance in expanded sensor placement configurations, while being unrestricted by the number of sensors of the volumetric capture system.
1 code implementation • 26 Feb 2021 • Sarthak Pati, Siddhesh P. Thakur, İbrahim Ethem Hamamcı, Ujjwal Baid, Bhakti Baheti, Megh Bhalerao, Orhun Güley, Sofia Mouchtaris, David Lang, Spyridon Thermos, Karol Gotkowski, Camila González, Caleb Grenko, Alexander Getka, Brandon Edwards, Micah Sheller, Junwen Wu, Deepthi Karkada, Ravi Panchumarthy, Vinayak Ahluwalia, Chunrui Zou, Vishnu Bashyam, Yuemeng Li, Babak Haghighi, Rhea Chitalia, Shahira Abousamra, Tahsin M. Kurc, Aimilia Gastounioti, Sezgin Er, Mark Bergman, Joel H. Saltz, Yong Fan, Prashant Shah, Anirban Mukhopadhyay, Sotirios A. Tsaftaris, Bjoern Menze, Christos Davatzikos, Despina Kontos, Alexandros Karargyris, Renato Umeton, Peter Mattson, Spyridon Bakas
Deep Learning (DL) has the potential to optimize machine learning in both the scientific and clinical communities.
4 code implementations • 27 Aug 2020 • Xiao Liu, Spyridon Thermos, Gabriele Valvano, Agisilaos Chartsias, Alison O'Neil, Sotirios A. Tsaftaris
In this paper, we conduct an empirical study to investigate the role of different biases in content-style disentanglement settings and unveil the relationship between the degree of disentanglement and task performance.
1 code implementation • ICCV 2019 • Vladimiros Sterzentsenko, Leonidas Saroglou, Anargyros Chatzitofis, Spyridon Thermos, Nikolaos Zioulis, Alexandros Doumanoglou, Dimitrios Zarpalas, Petros Daras
Specifically, the proposed autoencoder exploits multiple views of the same scene from different points of view in order to learn to suppress noise in a self-supervised end-to-end manner using depth and color information during training, yet only depth during inference.
1 code implementation • 6 Aug 2022 • Xiao Liu, Spyridon Thermos, Pedro Sanchez, Alison Q. O'Neil, Sotirios A. Tsaftaris
Maximisation of mutual information is achieved by introducing an auxiliary network and training with a latent regression loss.
1 code implementation • 26 Aug 2021 • Xiao Liu, Pedro Sanchez, Spyridon Thermos, Alison Q. O'Neil, Sotirios A. Tsaftaris
Disentangled representation learning has been proposed as an approach to learning general representations even in the absence of, or with limited, supervision.
2 code implementations • 24 Jun 2021 • Xiao Liu, Spyridon Thermos, Alison O'Neil, Sotirios A. Tsaftaris
We explicitly model the representations related to domain shifts.
1 code implementation • 29 Jun 2022 • Xiao Liu, Spyridon Thermos, Pedro Sanchez, Alison Q. O'Neil, Sotirios A. Tsaftaris
Moreover, with a reconstruction module, unlabeled data can also be used to learn the vMF kernels and likelihoods by recombining them to reconstruct the input image.
1 code implementation • 26 Aug 2020 • Xiao Liu, Spyridon Thermos, Agisilaos Chartsias, Alison O'Neil, Sotirios A. Tsaftaris
Robust cardiac image segmentation is still an open challenge due to the inability of the existing methods to achieve satisfactory performance on unseen data of different domains.
1 code implementation • 4 Jul 2021 • Spyridon Thermos, Xiao Liu, Alison O'Neil, Sotirios A. Tsaftaris
Motivated by the ability to disentangle images into spatial anatomy (tensor) factors and accompanying imaging (vector) representations, we propose a framework termed "disentangled anatomy arithmetic", in which a generative model learns to combine anatomical factors of different input images such that when they are re-entangled with the desired imaging modality (e. g. MRI), plausible new cardiac images are created with the target characteristics.
no code implementations • CVPR 2017 • Spyridon Thermos, Georgios Th. Papadopoulos, Petros Daras, Gerasimos Potamianos
It is well-established by cognitive neuroscience that human perception of objects constitutes a complex process, where object appearance information is combined with evidence about the so-called object "affordances", namely the types of actions that humans typically perform when interacting with them.
no code implementations • 18 Apr 2020 • Spyridon Thermos, Petros Daras, Gerasimos Potamianos
In particular, we design an autoencoder that is trained using ground-truth labels of only the last frame of the sequence, and is able to infer pixel-wise affordance labels in both videos and static images.
no code implementations • CVPR 2022 • Anargyros Chatzitofis, Georgios Albanis, Nikolaos Zioulis, Spyridon Thermos
Traditional marker-based motion capture requires excessive and specialized equipment, hindering accessibility and wider adoption.
no code implementations • 13 Jun 2023 • Xiao Liu, Pedro Sanchez, Spyridon Thermos, Alison Q. O'Neil, Sotirios A. Tsaftaris
By modelling the compositional representations with learnable von-Mises-Fisher (vMF) kernels, we explore how different design and learning biases can be used to enforce the representations to be more compositionally equivariant under un-, weakly-, and semi-supervised settings.
no code implementations • 25 Sep 2023 • Georgios Albanis, Nikolaos Zioulis, Spyridon Thermos, Anargyros Chatzitofis, Kostas Kolomvatsos
By relying on a unified representation, we show that training such a model is not bound to high-end MoCap training data acquisition, and exploit the advances in marker-less MoCap to acquire the necessary data.