Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Sameni, Sepehr"'
In this work we propose a novel method for unsupervised controllable video generation. Once trained on a dataset of unannotated videos, at inference our model is capable of both composing scenes of predefined object parts and animating them in a plau
Externí odkaz:
http://arxiv.org/abs/2403.14368
The growing interest in novel view synthesis, driven by Neural Radiance Field (NeRF) models, is hindered by scalability issues due to their reliance on precisely annotated multi-view images. Recent models address this by fine-tuning large text2image
Externí odkaz:
http://arxiv.org/abs/2312.04337
We propose Spatio-temporal Crop Aggregation for video representation LEarning (SCALE), a novel method that enjoys high scalability at both training and inference time. Our model builds long-range video features by learning from sets of video clip-lev
Externí odkaz:
http://arxiv.org/abs/2211.17042
We introduce a novel generative model for video prediction based on latent flow matching, an efficient alternative to diffusion-based models. In contrast to prior work, we keep the high costs of modeling the past during training and inference at bay
Externí odkaz:
http://arxiv.org/abs/2211.14575
In this paper, we introduce a novel self-supervised learning (SSL) loss for image representation learning. There is a growing belief that generalization in deep neural networks is linked to their ability to discriminate object shapes. Since object sh
Externí odkaz:
http://arxiv.org/abs/2204.04788
Autor:
Davtyan, Aram, Sameni, Sepehr, Cerkezi, Llukman, Meishvilli, Givi, Bielski, Adam, Favaro, Paolo
Optimization is often cast as a deterministic problem, where the solution is found through some iterative procedure such as gradient descent. However, when training neural networks the loss function changes over (iteration) time due to the randomized
Externí odkaz:
http://arxiv.org/abs/2107.03331
We introduce a novel generative model for video prediction based on latent flow matching, an efficient alternative to diffusion-based models. In contrast to prior work that either incurs a high training cost by modeling the past through a memory stat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::940bcd30acd019ed41643ad3c5e28e4e