Coarse-to-Fine Target Speaker Extraction Based on Contextual Information Exploitation.
Xue YangChangchun BaoXianhong ChenPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2024)
Keyphrases
- contextual information
- coarse to fine
- multiscale
- multiresolution
- context aware
- object detection
- image registration
- hierarchical segmentation
- image pyramid
- appearance information
- hierarchical representation
- contextual knowledge
- dynamic programming
- active shape model
- matching scheme
- spatial context
- high level
- information extraction
- global information
- multi view face detection
- computer vision
- deformable surface model
- image regions
- higher order
- pattern recognition