Zobrazeno 1 - 10
of 4 138
pro vyhledávání: '"Bozić, A."'
Progressing beyond centralized AI is of paramount importance, yet, distributed AI solutions, in particular various federated learning (FL) algorithms, are often not comprehensively assessed, which prevents the research community from identifying the
Externí odkaz:
http://arxiv.org/abs/2407.14154
Publikováno v:
WWW '24: Companion Proceedings of the ACM on Web Conference 2024 May 2024 Pages 1431 - 1440
While the use of machine learning for the detection of propaganda techniques in text has garnered considerable attention, most approaches focus on "black-box" solutions with opaque inner workings. Interpretable approaches provide a solution, however,
Externí odkaz:
http://arxiv.org/abs/2407.11827
Autor:
Božić, Matej, Horvat, Marko
This article presents a review of typical techniques used in three distinct aspects of deep learning model development for audio generation. In the first part of the article, we provide an explanation of audio representations, beginning with the fund
Externí odkaz:
http://arxiv.org/abs/2406.00146
Autor:
Condor, Jorge, Speierer, Sebastien, Bode, Lukas, Bozic, Aljaz, Green, Simon, Didyk, Piotr, Jarabo, Adrian
Efficient scene representations are essential for many computer graphics applications. A general unified representation that can handle both surfaces and volumes simultaneously, remains a research challenge. Inspired by recent methods for scene recon
Externí odkaz:
http://arxiv.org/abs/2405.15425
Autor:
Höllein, Lukas, Božič, Aljaž, Müller, Norman, Novotny, David, Tseng, Hung-Yu, Richardt, Christian, Zollhöfer, Michael, Nießner, Matthias
3D asset generation is getting massive amounts of attention, inspired by the recent success of text-guided 2D content creation. Existing text-to-3D methods use pretrained text-to-image diffusion models in an optimization problem or fine-tune them on
Externí odkaz:
http://arxiv.org/abs/2403.01807
Autor:
Fischer, Michael, Li, Zhengqin, Nguyen-Phuoc, Thu, Bozic, Aljaz, Dong, Zhao, Marshall, Carl, Ritschel, Tobias
A Neural Radiance Field (NeRF) encodes the specific relation of 3D geometry and appearance of a scene. We here ask the question whether we can transfer the appearance from a source NeRF onto a target 3D geometry in a semantically meaningful way, such
Externí odkaz:
http://arxiv.org/abs/2402.08622
Autor:
Wang, Ziyan, Nam, Giljoo, Bozic, Aljaz, Cao, Chen, Saragih, Jason, Zollhoefer, Michael, Hodgins, Jessica
Hair plays a significant role in personal identity and appearance, making it an essential component of high-quality, photorealistic avatars. Existing approaches either focus on modeling the facial region only or rely on personalized models, limiting
Externí odkaz:
http://arxiv.org/abs/2312.08679
This work presents an analysis of the effectiveness of using standard shallow feed-forward networks to mimic the behavior of the attention mechanism in the original Transformer model, a state-of-the-art architecture for sequence-to-sequence tasks. We
Externí odkaz:
http://arxiv.org/abs/2311.10642
Autor:
Xu, Linning, Agrawal, Vasu, Laney, William, Garcia, Tony, Bansal, Aayush, Kim, Changil, Bulò, Samuel Rota, Porzi, Lorenzo, Kontschieder, Peter, Božič, Aljaž, Lin, Dahua, Zollhöfer, Michael, Richardt, Christian
We present an end-to-end system for the high-fidelity capture, model reconstruction, and real-time rendering of walkable spaces in virtual reality using neural radiance fields. To this end, we designed and built a custom multi-camera rig to densely c
Externí odkaz:
http://arxiv.org/abs/2311.02542
Autor:
Tretschk, Edith, Golyanik, Vladislav, Zollhoefer, Michael, Bozic, Aljaz, Lassner, Christoph, Theobalt, Christian
Existing methods for the 4D reconstruction of general, non-rigidly deforming objects focus on novel-view synthesis and neglect correspondences. However, time consistency enables advanced downstream tasks like 3D editing, motion analysis, or virtual-a
Externí odkaz:
http://arxiv.org/abs/2308.08258