GenMM: Geometrically and Temporally Consistent Multimodal Data Generation for Video and LiDAR.
Bharat SinghViveka KulhariaLuyu YangAvinash RavichandranAmbrish TyagiAshish ShrivastavaPublished in: CoRR (2024)
Keyphrases
- temporally consistent
- data generation
- temporal consistency
- optical flow
- depth map
- superpixels
- active learning
- key frames
- motion parameters
- processing speed
- streaming data
- data streams
- video data
- temporal constraints
- video streams
- multimedia
- motion model
- image sequences
- point cloud
- high throughput
- disparity map
- stereo matching
- sliding window
- co training
- motion estimation
- high resolution
- pairwise
- multiscale