Controllable Length Control Neural Encoder-Decoder via Reinforcement Learning.
Junyi BianBaojun LinKe ZhangZhaohui YanHong TangYonghe ZhangPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- video codec
- low complexity
- distributed video coding
- control problems
- decoding process
- optimal control
- wyner ziv video coding
- error control
- robot control
- fitted q iteration
- neural network
- control policy
- network architecture
- state space
- rate distortion
- wyner ziv
- mpeg avc
- control system
- distributed source coding
- motion estimation
- successive approximation
- machine learning
- action selection
- function approximation
- markov decision processes
- motion compensated prediction
- video coding scheme
- noisy channel