SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition.
Sourav GargMichael MilfordPublished in: CoRR (2021)
Keyphrases
- image sequences
- depth map
- point cloud
- spatio temporal
- video sequences
- motion analysis
- structure from motion
- motion estimation
- infrared
- motion parameters
- optical flow
- structured light
- image frames
- optic flow
- feature points
- motion segmentation
- consecutive images
- moving camera
- three dimensional
- image motion
- background subtraction
- motion field
- computer vision
- apparent motion
- automatic tracking
- consecutive frames
- least squares
- camera motion
- motion model
- flow field
- model selection
- super resolution
- motion detection
- velocity field
- image data
- human motion
- point sets