Zobrazeno 1 - 10
of 58
pro vyhledávání: '"Oldfield, James"'
Autor:
Oldfield, James, Georgopoulos, Markos, Chrysos, Grigorios G., Tzelepis, Christos, Panagakis, Yannis, Nicolaou, Mihalis A., Deng, Jiankang, Patras, Ioannis
The Mixture of Experts (MoE) paradigm provides a powerful way to decompose dense layers into smaller, modular computations often more amenable to human interpretation, debugging, and editability. However, a major challenge lies in the computational c
Externí odkaz:
http://arxiv.org/abs/2402.12550
Autor:
Oldfield, James, Tzelepis, Christos, Panagakis, Yannis, Nicolaou, Mihalis A., Patras, Ioannis
Latent image representations arising from vision-language models have proved immensely useful for a variety of downstream tasks. However, their utility is limited by their entanglement with respect to different visual attributes. For instance, recent
Externí odkaz:
http://arxiv.org/abs/2305.14053
This work addresses the problem of discovering non-linear interpretable paths in the latent space of pre-trained GANs in a model-agnostic manner. In the proposed method, the discovery is driven by a set of pairs of natural language sentences with con
Externí odkaz:
http://arxiv.org/abs/2206.02104
Autor:
Oldfield, James, Tzelepis, Christos, Panagakis, Yannis, Nicolaou, Mihalis A., Patras, Ioannis
Recent advances in the understanding of Generative Adversarial Networks (GANs) have led to remarkable progress in visual editing and synthesis tasks, capitalizing on the rich semantics that are embedded in the latent spaces of pre-trained GANs. Howev
Externí odkaz:
http://arxiv.org/abs/2206.00048
Generative Adversarial Networks (GANs) are the driving force behind the state-of-the-art in image generation. Despite their ability to synthesize high-resolution photo-realistic images, generating content with on-demand conditioning of different gran
Externí odkaz:
http://arxiv.org/abs/2112.12911
Autor:
Oldfield, James, Georgopoulos, Markos, Panagakis, Yannis, Nicolaou, Mihalis A., Patras, Ioannis
This paper addresses the problem of finding interpretable directions in the latent space of pre-trained Generative Adversarial Networks (GANs) to facilitate controllable image synthesis. Such interpretable directions correspond to transformations tha
Externí odkaz:
http://arxiv.org/abs/2111.11736
Autor:
Panagakis, Yannis, Kossaifi, Jean, Chrysos, Grigorios G., Oldfield, James, Nicolaou, Mihalis A., Anandkumar, Anima, Zafeiriou, Stefanos
Tensors, or multidimensional arrays, are data structures that can naturally represent visual data of multiple dimensions. Inherently able to efficiently capture structured, latent semantic spaces and high-order interactions, tensors have a long histo
Externí odkaz:
http://arxiv.org/abs/2107.03436
A significant limiting factor in training fair classifiers relates to the presence of dataset bias. In particular, face datasets are typically biased in terms of attributes such as gender, age, and race. If not mitigated, bias leads to algorithms tha
Externí odkaz:
http://arxiv.org/abs/2006.03985
Recently, a multitude of methods for image-to-image translation have demonstrated impressive results on problems such as multi-domain or multi-attribute transfer. The vast majority of such works leverages the strengths of adversarial learning and dee
Externí odkaz:
http://arxiv.org/abs/1904.04772
Autor:
Oldfield, James Peter
Thesis advisor: Jeffrey Bloechl
The idea of moral action seems to contain a paradox. On the one hand, it seems that in performing such an act one is obligated, bound to the act by something external. On the other hand, it seems that such an act
The idea of moral action seems to contain a paradox. On the one hand, it seems that in performing such an act one is obligated, bound to the act by something external. On the other hand, it seems that such an act
Externí odkaz:
http://hdl.handle.net/2345/bc-ir:105055