DC-EDN: densely connected encoder-decoder network with reinforced depthwise convolution for face alignment.
Lianping YangHongliang ZhangPanpan WeiYubo SunXiangde ZhangPublished in: Appl. Intell. (2021)
Keyphrases
- face alignment
- densely connected
- sparsely connected
- community structure
- image processing
- active appearance models
- multi view
- distributed video coding
- information networks
- network structure
- machine learning
- facial images
- community detection
- active shape model
- real world networks
- motion estimation
- face detection
- complex networks
- bit rate
- active learning
- multiscale