TiV-ODE: A Neural ODE-based Approach for Controllable Video Generation From Text-Image Pairs.
Yucheng XuNanbo LiArushi GoelZonghai YaoZijian GuoHamidreza KasaeiMohammadreza KasaeiZhibin LiPublished in: ICRA (2024)
Keyphrases
- image pairs
- ordinary differential equations
- stereo images
- image registration
- single image
- text generation
- epipolar geometry
- feature matching
- image matching
- neural network
- video data
- fundamental matrix
- multimedia
- video frames
- input image
- disparity map
- video sequences
- dense correspondences
- real time
- differential equations
- geometric transformations
- wide baseline
- registration process
- space time
- dynamic scenes
- pattern recognition
- video content
- image correspondences
- computer vision
- multi view