Long-Term Temporal Convolutions for Action Recognition

作者：Gül Varol, Ivan Laptev, Cordelia Schmid · 发表于：IEEE Transactions on Pattern Analysis and Machine Intelligence · 年份：2017 · DOI：10.1109/tpami.2017.2712608 · 被引用次数：953 · 研究领域：Human Pose and Action Recognition、Anomaly Detection Techniques and Applications、Multimodal Machine Learning Applications

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%).