Temporal Gaussian Mixture Layer for Videos

Aj Piergiovanni, Michael Ryoo
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:5152-5161, 2019.

Abstract

We introduce a new convolutional layer named the Temporal Gaussian Mixture (TGM) layer and present how it can be used to efficiently capture longer-term temporal information in continuous activity videos. The TGM layer is a temporal convolutional layer governed by a much smaller set of parameters (e.g., location/variance of Gaussians) that are fully differentiable. We present our fully convolutional video models with multiple TGM layers for activity detection. The extensive experiments on multiple datasets, including Charades and MultiTHUMOS, confirm the effectiveness of TGM layers, significantly outperforming the state-of-the-arts.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-piergiovanni19a, title = {Temporal {G}aussian Mixture Layer for Videos}, author = {Piergiovanni, Aj and Ryoo, Michael}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {5152--5161}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/piergiovanni19a/piergiovanni19a.pdf}, url = {https://proceedings.mlr.press/v97/piergiovanni19a.html}, abstract = {We introduce a new convolutional layer named the Temporal Gaussian Mixture (TGM) layer and present how it can be used to efficiently capture longer-term temporal information in continuous activity videos. The TGM layer is a temporal convolutional layer governed by a much smaller set of parameters (e.g., location/variance of Gaussians) that are fully differentiable. We present our fully convolutional video models with multiple TGM layers for activity detection. The extensive experiments on multiple datasets, including Charades and MultiTHUMOS, confirm the effectiveness of TGM layers, significantly outperforming the state-of-the-arts.} }
Endnote
%0 Conference Paper %T Temporal Gaussian Mixture Layer for Videos %A Aj Piergiovanni %A Michael Ryoo %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-piergiovanni19a %I PMLR %P 5152--5161 %U https://proceedings.mlr.press/v97/piergiovanni19a.html %V 97 %X We introduce a new convolutional layer named the Temporal Gaussian Mixture (TGM) layer and present how it can be used to efficiently capture longer-term temporal information in continuous activity videos. The TGM layer is a temporal convolutional layer governed by a much smaller set of parameters (e.g., location/variance of Gaussians) that are fully differentiable. We present our fully convolutional video models with multiple TGM layers for activity detection. The extensive experiments on multiple datasets, including Charades and MultiTHUMOS, confirm the effectiveness of TGM layers, significantly outperforming the state-of-the-arts.
APA
Piergiovanni, A. & Ryoo, M.. (2019). Temporal Gaussian Mixture Layer for Videos. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:5152-5161 Available from https://proceedings.mlr.press/v97/piergiovanni19a.html.

Related Material