A Spatial and Temporal Features Mixture Model with Body Parts for Video-based Person Re-Identification.
Jie LiuCheng SunXiang XuBaomin XuShuangyuan YuPublished in: CoRR (2018)
Keyphrases
- spatial and temporal
- body parts
- mixture model
- spatio temporal
- human body
- video frames
- gaussian mixture model
- space time
- em algorithm
- human motion
- action recognition
- generative model
- model selection
- probabilistic model
- pose estimation
- expectation maximization
- feature vectors
- feature extraction
- image features
- video sequences
- maximum likelihood
- prior knowledge
- visual features
- machine learning
- reinforcement learning
- training data
- language model