ML Anthology

Long-Term Feature Banks for Detailed Video Understanding

Chao-Yuan Wu, Christoph Feichtenhofer, Haoqi Fan, Kaiming He, Philipp Krahenbuhl, Ross Girshick
CVPR 2019
doi:10.1109/CVPR.2019.00037 /cvpr/2019/wu2019cvpr-longterm/

Abstract

To understand the world, we humans constantly need to relate the present to the past, and put events in context. In this paper, we enable existing video models to do the same. We propose a long-term feature bank--supportive information extracted over the entire span of a video--to augment state-of-the-art video models that otherwise would only view short clips of 2-5 seconds. Our experiments demonstrate that augmenting 3D convolutional networks with a long-term feature bank yields state-of-the-art results on three challenging video datasets: AVA, EPIC-Kitchens, and Charades. Code is available online.

PDF CVPR Semantic Scholar

Cite

Text

Wu et al. "Long-Term Feature Banks for Detailed Video Understanding." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019. doi:10.1109/CVPR.2019.00037

Markdown

[Wu et al. "Long-Term Feature Banks for Detailed Video Understanding." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019.](https://mlanthology.org/cvpr/2019/wu2019cvpr-longterm/) doi:10.1109/CVPR.2019.00037

BibTeX

@inproceedings{wu2019cvpr-longterm,
  title     = {{Long-Term Feature Banks for Detailed Video Understanding}},
  author    = {Wu, Chao-Yuan and Feichtenhofer, Christoph and Fan, Haoqi and He, Kaiming and Krahenbuhl, Philipp and Girshick, Ross},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2019},
  doi       = {10.1109/CVPR.2019.00037},
  url       = {https://mlanthology.org/cvpr/2019/wu2019cvpr-longterm/}
}