Arbitrary-shaped scene text detection with keypoint-based shape representation.
Shuxin QinLin ChenPublished in: Int. J. Document Anal. Recognit. (2022)
Keyphrases
- shape representation
- text detection
- arbitrary shaped
- urban scenes
- outdoor scenes
- shape analysis
- multiscale
- shape matching
- computationally efficient
- shape descriptors
- video analysis
- d scene
- keypoints
- object recognition
- single image
- video sequences
- matching algorithm
- image sequences
- range data
- vanishing points
- complex scenes
- arbitrary shape
- three dimensional
- shape prior
- viewpoint
- coarse to fine
- computer vision
- point cloud
- photorealistic
- image based rendering
- input image