MEDA: Multi-output Encoder-Decoder for Spatial Attention in Convolutional Neural Networks.
Huayu LiAbolfazl RaziPublished in: ACSSC (2019)
Keyphrases
- convolutional neural networks
- temporal correlation
- distributed video coding
- video codec
- decoding process
- low complexity
- error control
- wyner ziv video coding
- spatial correlation
- spatial and temporal
- distributed source coding
- noisy channel
- convolutional network
- motion estimation
- spatial information
- rate distortion
- wyner ziv
- bit rate
- video coding scheme
- video coding
- discriminative power
- rate allocation
- shift invariant
- successive approximation
- mpeg avc
- wavelet transform
- inter frame
- pixel domain
- transform domain
- video encoder
- error concealment