Beyond Keypoint Coding: Temporal Evolution Inference with Compact Feature Representation for Talking Face Video Compression.
Bolin ChenZhao WangBinzhe LiRongqun LinShiqi WangYan YePublished in: DCC (2022)
Keyphrases
- video compression
- feature representation
- temporal evolution
- image and video compression
- motion compensation
- video coding
- temporal redundancy
- motion estimation
- distributed video coding
- keypoints
- feature descriptors
- face recognition
- face images
- video data
- motion compensated
- feature extraction
- temporal information
- non stationary
- compression ratio
- low dimensional
- coding method
- inter frame
- sparse representation
- coding scheme
- facial expressions
- feature set
- bit rate
- image patches
- motion vectors
- image quality
- multiscale
- video sequences
- image processing
- multiresolution
- image features
- local binary pattern
- natural images