English

Long-term Temporal Convolutions for Action Recognition

Computer Vision and Pattern Recognition 2017-06-05 v2

Abstract

Typical human actions last several seconds and exhibit characteristic spatio-temporal structure. Recent methods attempt to capture this structure and learn action representations with convolutional neural networks. Such representations, however, are typically learned at the level of a few video frames failing to model actions at their full temporal extent. In this work we learn video representations using neural networks with long-term temporal convolutions (LTC). We demonstrate that LTC-CNN models with increased temporal extents improve the accuracy of action recognition. We also study the impact of different low-level representations, such as raw values of video pixels and optical flow vector fields and demonstrate the importance of high-quality optical flow estimation for learning accurate action models. We report state-of-the-art results on two challenging benchmarks for human action recognition UCF101 (92.7%) and HMDB51 (67.2%).

Keywords

Cite

@article{arxiv.1604.04494,
  title  = {Long-term Temporal Convolutions for Action Recognition},
  author = {Gül Varol and Ivan Laptev and Cordelia Schmid},
  journal= {arXiv preprint arXiv:1604.04494},
  year   = {2017}
}
R2 v1 2026-06-22T13:33:19.050Z