A novel no-sensors 3D model reconstruction from monocular video frames for a dynamic environment.
Ghada M. FathyHanan A. HassanWalaa M. ShetaFatma A. OmaraEmad NabilPublished in: PeerJ Comput. Sci. (2021)
Keyphrases
- dynamic environments
- video frames
- visual slam
- video sequences
- sensory information
- video data
- video segmentation
- image sequences
- path planning
- consecutive frames
- sensor data
- video content
- video clips
- temporal coherence
- successive frames
- video images
- single frame
- changing environment
- multi sensor
- pose estimation
- adjacent frames
- input video
- real environment
- potential field
- mobile robot
- real time
- monocular video sequences
- foreground objects
- camera movement
- vision system
- metadata