Recognition of multiple objects based on global image consistency between a scene hypothesis and the acquired image.
Manabu HashimotoShin'ichi KurodaKazuhiko SumiTeruo UsamiShuji NakataPublished in: Systems and Computers in Japan (2001)
Keyphrases
- multiple objects
- input image
- complex scenes
- single image
- image features
- occluded objects
- acquired images
- object recognition
- individual objects
- image noise
- image matching
- scene understanding
- multiple images
- occlusion handling
- cluttered scenes
- object models
- image processing
- image regions
- high resolution
- multi object
- super resolution
- dynamic programming
- search algorithm
- optimal solution