Файл:Unsupervised Learningof Video Representation susing LSTMs 1502.04681.pdf

Материал из Материалы по машинному обучению
Перейти к: навигация, поиск
Unsupervised_Learningof_Video_Representation_susing_LSTMs_1502.04681.pdf(0 × 0 пикселей, размер файла: 2,19 МБ, MIME-тип: application/pdf)
  • Nitish Srivastava NITISH@CS.TORONTO.EDU
  • Ruslan Salakhutdinov RSALAKHU@CS.TORONTO.EDU
  • University of Toronto, 6 Kings College Road, Toronto, ON M5S 3G4 CANADA


We use multilayer Long Short Term Memory (LSTM) networks to learn representations of video sequences. Our model uses an encoder LSTM to map an input sequence into a fixed length representation. This representation is decoded using single or multiple decoder LSTMs to perform different tasks, such as reconstructing the input sequence, or predicting the future sequence. We experiment with two kinds of input sequences – patches of image pixels and high-level representations (“percepts”) of video framesextractedusingapretrainedconvolutional net. We explore different design choices such as whether the decoder LSTMs should condition on the generated output. We analyze the outputs of the model qualitatively to see how well the model can extrapolate the learned video representation into the future and into the past. We try to visualize and interpret the learned features. We stress test the model by running it on longer time scales and on out-of-domain data. We further evaluate the representations by finetuning them for a supervised learning problem – human action recognition on the UCF-101 and HMDB-51 datasets. We show that the representations help improve classification accuracy, especially when there are only a few training examples. Even models pretrained on unrelated datasets (300 hours of YouTube videos) can help action recognition performance.

История файла

Нажмите на дату/время, чтобы просмотреть, как тогда выглядел файл.

текущий11:55, 22 декабря 20160 × 0 (2,19 МБ)Slikos (обсуждение | вклад)
  • Вы не можете перезаписать этот файл.

Следующая 1 страница ссылается на данный файл: