Coarse-to-Fine Dual Encoders are Better Frame Identification Learners.
Kaikai AnCe ZhengBofei GaoHaozhe ZhaoBaobao ChangPublished in: CoRR (2023)
Keyphrases
- coarse to fine
- successive frames
- multiscale
- multiresolution
- object detection
- feature correspondences
- hierarchical segmentation
- image registration
- video frames
- learning environment
- e learning
- active shape model
- matching scheme
- hierarchical representation
- image pyramid
- image processing
- dynamic programming
- scan line
- shape model
- deformable surface model
- optical flow estimation
- machine learning
- pattern recognition
- computer vision