Action Recognition Using Hybrid Feature Descriptor and VLAD Video Encoding.
Dong XingXianzhong WangHongtao LuPublished in: ACCV Workshops (1) (2014)
Keyphrases
- feature descriptors
- video encoding
- low complexity
- local binary pattern
- low bit rate
- image features
- computational power
- feature points
- motion estimation
- shape descriptors
- image matching
- video encoder
- texture classification
- reduce the computational complexity
- shape features
- computationally intensive
- resource constrained
- distance metric
- video coding
- machine learning
- image registration
- multiresolution
- video sequences
- digital video
- multiscale
- face recognition