Learning Light Field Synthesis with Multi-Plane Images: Scene Encoding as a Recurrent Segmentation Task.
Tomás VölkerGuillaume BoissonBertrand ChupeauPublished in: ICIP (2020)
Keyphrases
- three dimensional
- acquired images
- light field
- epipolar plane
- planar regions
- multiple images
- image based rendering
- input image
- scene depth
- scene geometry
- d scene
- outdoor scenes
- complex scenes
- captured images
- single image
- light field rendering
- vanishing points
- image segmentation
- static scene
- multi view
- d objects
- structure from motion
- video data
- view dependent
- densely sampled
- single shot
- level set
- face recognition across pose