Robust Semi-Automatic Depth Map Generation in Unconstrained Images and Video Sequences for 2D to Stereoscopic 3D Conversion.
Raymond PhanDimitrios AndroutsosPublished in: IEEE Trans. Multim. (2014)
Keyphrases
- semi automatic
- depth map
- images and video sequences
- stereo pair
- stereoscopic images
- depth information
- three dimensional
- fully automatic
- multi view
- gold standard
- d scene
- stereo matching
- depth cues
- high quality
- image sequences
- high resolution
- depth estimation
- motion parallax
- object boundaries
- input image
- view synthesis
- landmark extraction
- low resolution
- real scenes
- input data
- view dependent
- video sequences
- stereoscopic video
- shape from focus
- depth image based rendering
- ontology construction
- ground truth
- face recognition