An Encoder-Decoder Based Audio Captioning System with Transfer and Reinforcement Learning.
Xinhao MeiQiushi HuangXubo LiuGengyun ChenJingqian WuYusong WuJinzheng ZhaoShengchen LiTom KoH. TangXi ShaoMark D. PlumbleyWenwu WangPublished in: DCASE (2021)
Keyphrases
- reinforcement learning
- video codec
- decoding process
- distributed video coding
- error control
- transfer learning
- low complexity
- noisy channel
- mpeg avc
- wyner ziv
- rate distortion
- wyner ziv video coding
- video coding scheme
- successive approximation
- video coding
- multimedia
- function approximation
- distributed source coding
- turbo codes
- bit rate
- visual information
- state space
- error resilience
- audio visual
- pixel domain
- video signals
- motion estimation
- low bit rate
- learning algorithm
- transform domain
- video compression
- error correction
- bit budget
- knowledge transfer
- video quality
- vector quantization
- motion vectors
- motion compensated prediction
- rate allocation
- temporal correlation
- error concealment
- video transcoding
- video coder
- markov decision processes
- video coding standard
- inter frame
- image coding
- video encoder
- previously learned
- decoding algorithm
- compression efficiency
- error propagation
- compressed video
- motion compensation
- rate control