VideoMV: Consistent Multi-View Generation Based on Large Video Generative Model.
Qi ZuoXiaodong GuLingteng QiuYuan DongZhengyi ZhaoWeihao YuanRui PengSiyu ZhuZilong DongLiefeng BoQixing HuangPublished in: CoRR (2024)
Keyphrases
- generative model
- multi view
- semi supervised
- dynamic scenes
- free viewpoint
- single view
- depth video
- multiple views
- bayesian framework
- probabilistic model
- depth map
- dynamic textures
- three dimensional
- video data
- video sequences
- d objects
- prior knowledge
- em algorithm
- video streams
- video content
- stereoscopic video
- view synthesis
- co training
- topic models
- object categories
- latent dirichlet allocation
- light field
- video frames
- expectation maximization
- decision trees
- supervised learning
- gaussian process
- multi view learning
- labeled data
- depth image based rendering