Recovering Relative Depth from Low-Level Features Without Explicit T-junction Detection and Interpretation.
Felipe CaldereroVicent CasellesPublished in: Int. J. Comput. Vis. (2013)
Keyphrases
- low level features
- junction detection
- relative depth
- high level
- low level
- higher level
- biologically motivated
- visual features
- high level semantics
- relevance feedback
- image retrieval
- image sequences
- image representation
- saliency map
- semantic information
- image velocity
- object recognition
- motion parameters
- key frames
- image structure
- motion field
- multiscale
- ego motion
- visual information
- image content
- least squares
- domain knowledge
- image processing