TON-VIO: Online Time Offset Modeling Networks for Robust Temporal Alignment in High Dynamic Motion VIO.
Chaoran XiongGuoqing LiuQi WuSongpengcheng XiaTong HuaKehui MaZhen SunYan XiangLing PeiPublished in: CoRR (2024)
Keyphrases
- optical flow
- image sequences
- human motion
- motion analysis
- space time
- motion estimation
- motion detection
- long video sequences
- high robustness
- dynamic networks
- motion field
- social networks
- online learning
- motion tracking
- motion model
- dynamic environments
- wide range
- motion segmentation
- robust estimation
- robust tracking
- linear dynamical systems
- network structure
- dynamic scenes
- motion planning
- optical flow estimation
- computationally efficient
- echo state networks