Zobrazeno 1 - 10
of 10
pro vyhledávání: '"Huang, Xingchang"'
Classical generative diffusion models learn an isotropic Gaussian denoising process, treating all spatial regions uniformly, thus neglecting potentially valuable structural information in the data. Inspired by the long-established work on anisotropic
Externí odkaz:
http://arxiv.org/abs/2410.01540
We introduce a theoretical and practical framework for efficient importance sampling of mini-batch samples for gradient estimation from single and multiple probability distributions. To handle noisy gradients, our framework dynamically evolves the im
Externí odkaz:
http://arxiv.org/abs/2407.15525
Autor:
Huang, Xingchang, Salaün, Corentin, Vasconcelos, Cristina, Theobalt, Christian, Öztireli, Cengiz, Singh, Gurprit
Most of the existing diffusion models use Gaussian noise for training and sampling across all time steps, which may not optimally account for the frequency contents reconstructed by the denoising network. Despite the diverse applications of correlate
Externí odkaz:
http://arxiv.org/abs/2402.04930
Machine learning problems rely heavily on stochastic gradient descent (SGD) for optimization. The effectiveness of SGD is contingent upon accurately estimating gradients from a mini-batch of data samples. Instead of the commonly used uniform sampling
Externí odkaz:
http://arxiv.org/abs/2311.14468
Publikováno v:
ACM Transactions on Graphics 2023 Volume 42 Issue 4 Article No.: 53
Point patterns are characterized by their density and correlation. While spatial variation of density is well-understood, analysis and synthesis of spatially-varying correlation is an open challenge. No tools are available to intuitively edit such po
Externí odkaz:
http://arxiv.org/abs/2308.10517
Publikováno v:
NeurIPS 2020
Existing research on continual learning of a sequence of tasks focused on dealing with catastrophic forgetting, where the tasks are assumed to be dissimilar and have little shared knowledge. Some work has also been done to transfer previously learned
Externí odkaz:
http://arxiv.org/abs/2112.10017
Autor:
Huang, Xingchang1 (AUTHOR), Memari, Pooran2 (AUTHOR), Seidel, Hans‐Peter1 (AUTHOR), Singh, Gurprit1 (AUTHOR)
Publikováno v:
Computer Graphics Forum. Jul2022, Vol. 41 Issue 4, p169-179. 11p. 7 Color Photographs, 3 Black and White Photographs, 5 Charts, 3 Graphs.
Autor:
Kim, Byungsoo1 (AUTHOR), Huang, Xingchang1,2 (AUTHOR), Wuelfroth, Laura1 (AUTHOR), Tang, Jingwei1 (AUTHOR), Cordonnier, Guillaume1,3 (AUTHOR), Gross, Markus1 (AUTHOR), Solenthaler, Barbara1 (AUTHOR)
Publikováno v:
Computer Graphics Forum. May2022, Vol. 41 Issue 2, p97-110. 14p. 17 Color Photographs, 4 Black and White Photographs, 2 Diagrams, 1 Chart.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.