Spatially and Temporally Structured Global to Local Aggregation of Dynamic Depth Information for Action Recognition.
Yonghong HouShuang WangPichao WangZhimin GaoWanqing LiPublished in: IEEE Access (2018)
Keyphrases
- action recognition
- depth information
- depth sensors
- depth cameras
- depth map
- bag of words
- human actions
- activity recognition
- stereo vision
- depth images
- depth data
- action classification
- computer vision
- stereo camera
- scene flow
- body parts
- depth estimation
- human activities
- human pose
- recognizing human actions
- disparity map
- spatio temporal
- machine learning
- stereo matching
- global features
- multiple views
- human body
- multi view
- feature extraction
- action recognition in videos