A spatio-temporal interest point detector based on vorticity for action recognition.
Yuanbo ChenZhixuan LiXin GuoYanyun ZhaoAnni CaiPublished in: ICME Workshops (2013)
Keyphrases
- action recognition
- interest point detectors
- spatio temporal
- human actions
- velocity field
- flow field
- action recognition in videos
- scale invariant
- image sequences
- vector field
- view invariant
- optical flow
- recognition of human actions
- object recognition
- activity recognition
- computer vision
- bag of words
- motion field
- moving objects
- space time
- bag of features
- motion segmentation
- human activities
- keypoints
- scale space
- human motion
- affine invariant
- machine learning