We address the problem of detecting actions, such as drinking or opening a
door, in hours of challenging video data. We propose a model based on a
sequence of atomic action units, termed ''actoms'', that are characteristic for
the action. Our model represents the temporal structure of actions as a
sequence of histograms of actom-anchored visual features. Our representation,
which can be seen as a temporally structured extension of the bag-of-features,
is flexible, sparse and discriminative. We refer to our model as Actom Sequence
Model (ASM). Training requires the annotation of actoms for action clips. At
test time, actoms are detected automatically, based on a non-parametric model
of the distribution of actoms, which also acts as a prior on an action's
temporal structure. We present experimental results on two recent benchmarks
for temporal action detection. We show that our ASM method outperforms the
current state of the art in temporal action detection.
Below are some examples of actions correctly detected with our ASM approach.
The videos are rendered using the HTML5 video tag. To view the sequence of the
central frames of the actoms, just put your mouse pointer over the images (uses
Javascript).
Detection window
Central frames of actoms
An automatic "action summary" of the whole "Coffee and Cigarettes" test
sequence for "Drinking" is available below and
here.
