SFGAN: Unsupervised Generative Adversarial Learning of 3D Scene Flow from the 3D Scene Self.
Guangming WangChaokang JiangZehang ShenYanzi MiaoHesheng WangPublished in: Adv. Intell. Syst. (2022)
Keyphrases
- data sets
- d scene
- depth map
- single image
- unsupervised learning
- flow estimation
- camera parameters
- scene understanding
- optical flow
- image based rendering
- planar surfaces
- indoor scenes
- photorealistic
- coordinate frame
- learning algorithm
- camera viewpoint
- scene geometry
- scene structure
- ego motion
- learning process
- training set
- view synthesis
- depth information
- scene flow
- camera positions
- generative model
- view dependent
- least squares
- rigidly moving