A Super-Resolution-Based Feature Map Compression for Machine-Oriented Video Coding.
Jung Heum KangMuhammad Salman AliHye-Won JeongChang-Kyun ChoiYounhee KimSe-Yoon JeongSung-Ho BaeHui Yong KimPublished in: IEEE Access (2023)
Keyphrases
- super resolution
- video coding
- feature maps
- motion estimation
- low resolution
- rate distortion
- motion compensation
- motion vectors
- video compression
- high resolution
- image reconstruction
- bit rate
- video quality
- video codec
- multi frame
- self organizing maps
- image sequences
- video sequences
- feature vectors
- inter frame
- compression ratio
- high quality
- macroblock
- motion field
- spatial domain
- saliency map
- kernel function
- image quality
- image compression
- feature space
- single frame
- compression algorithm
- neural network
- optical flow
- computational complexity
- distributed video coding
- depth map
- reconstructed image
- visual quality
- data points
- object recognition
- three dimensional