Existing research on action recognition treats activities as monolithic events occurring in
videos. Recently, the benefits of formulating actions as a combination of atomic-actions have …

We describe a novel cross-modal embedding space for actions, named Action2Vec, which
combines linguistic cues from class labels with spatio-temporal features derived from video …

Despite the success of deep learning for static image understanding, it remains unclear what
are the most effective network architectures for spatial-temporal modeling in videos. In this …

Most action recognition models today are highly parameterized, and evaluated on datasets
with appearance-wise distinct classes. It has also been shown that 2D Convolutional Neural …

In egocentric videos, actions occur in quick succession. We capitalise on the action's
temporal context and propose a method that learns to attend to surrounding actions in order …

Much research on human action recognition has been oriented toward the performance gain
on lab-collected datasets. Yet real-world videos are more diverse, with more complicated …

We propose a self-supervised learning method to jointly reason about spatial and temporal
context for video recognition. Recent self-supervised approaches have used spatial context …

Action recognition in videos is a challenging task due to the complexity of the spatio-
temporal patterns to model and the difficulty to acquire and learn on large quantities of video …

Efficiently modeling spatial–temporal information in videos is crucial for action recognition.
To achieve this goal, state-of-the-art methods typically employ the convolution operator and …

Deep models have the risk of utilizing spurious clues to make predictions, eg, recognizing
actions via classifying the background scene. This problem severely degrades the open-set …