Zobrazeno 31 - 40
of 197
pro vyhledávání: '"Alayrac P"'
Autor:
Miech, Antoine, Zhukov, Dimitri, Alayrac, Jean-Baptiste, Tapaswi, Makarand, Laptev, Ivan, Sivic, Josef
Learning text-video embeddings usually requires a dataset of video clips with manually provided captions. However, such datasets are expensive and time consuming to create and therefore difficult to obtain on a large scale. In this work, we propose i
Externí odkaz:
http://arxiv.org/abs/1906.03327
Autor:
Uesato, Jonathan, Alayrac, Jean-Baptiste, Huang, Po-Sen, Stanforth, Robert, Fawzi, Alhussein, Kohli, Pushmeet
Recent work has uncovered the interesting (and somewhat surprising) finding that training models to be invariant to adversarial perturbations requires substantially larger datasets than those required for standard classification. This result is a key
Externí odkaz:
http://arxiv.org/abs/1905.13725
Autor:
Zhukov, Dimitri, Alayrac, Jean-Baptiste, Cinbis, Ramazan Gokberk, Fouhey, David, Laptev, Ivan, Sivic, Josef
In this paper we investigate learning visual models for the steps of ordinary tasks using weak supervision via instructional narrations and an ordered list of steps instead of strong supervision via temporal annotations. At the heart of our approach
Externí odkaz:
http://arxiv.org/abs/1903.08225
True video understanding requires making sense of non-lambertian scenes where the color of light arriving at the camera sensor encodes information about not just the last object it collided with, but about multiple mediums -- colored windows, dirty m
Externí odkaz:
http://arxiv.org/abs/1812.01461
Automatic generation of textual video descriptions that are time-aligned with video content is a long-standing goal in computer vision. The task is challenging due to the difficulty of bridging the semantic gap between the visual and natural language
Externí odkaz:
http://arxiv.org/abs/1809.08381
Spatio-temporal action detection in videos is typically addressed in a fully-supervised setup with manual annotation of training videos required at every frame. Since such annotation is extremely tedious and prohibits scalability, there is a clear ne
Externí odkaz:
http://arxiv.org/abs/1806.11328
Autor:
Vanessa Alayrac-Fielding
Publikováno v:
Miranda: Revue Pluridisciplinaire du Monde Anglophone, Vol 27 (2023)
Externí odkaz:
https://doaj.org/article/cdf0cea2162f4110851be993f7b02e08
Discriminative clustering has been successfully applied to a number of weakly-supervised learning tasks. Such applications include person and action recognition, text-to-video alignment, object co-segmentation and colocalization in videos and images.
Externí odkaz:
http://arxiv.org/abs/1707.09074
We propose SEARNN, a novel training algorithm for recurrent neural networks (RNNs) inspired by the "learning to search" (L2S) approach to structured prediction. RNNs have been widely successful in structured prediction applications such as machine tr
Externí odkaz:
http://arxiv.org/abs/1706.04499
Many human activities involve object manipulations aiming to modify the object state. Examples of common state changes include full/empty bottle, open/closed door, and attached/detached car wheel. In this work, we seek to automatically discover the s
Externí odkaz:
http://arxiv.org/abs/1702.02738