Visual Place Recognition
Visual Place Recognition is a task in the field of computer vision that aims to recognize locations by matching different views of the same place taken at different times. This technology is of great value in applications such as robot navigation, autonomous driving, and augmented reality, as it can enhance the environmental perception capabilities and positioning accuracy of systems.
Pittsburgh-30k-test
Patch-NetVLAD
Mapillary val
Patch-NetVLAD
St Lucia
MixVPR
Tokyo247
SelaVPR
Nordland
SelaVPR
Pittsburgh-250k-test
DINOv2 SALAD
Mapillary test
BoQ
17 Places
Baidu Mall
Nardo-Air R
AnyLoc-VLAD-DINO
Gardens Point
Hawkins
Laurel Caverns
Mid-Atlantic Ridge
Nardo-Air
Oxford RobotCar Dataset
VP-Air
AmsterTime
KITTI360pose
MambaPlace
Berlin Kudamm
HEAPUtil
Eynsham
BoQ
SF-XL test v1
EigenPlaces
SF-XL test v2
CosPlace
SPED
CV-Cities
MSLS
ProGEO
Oxford RobotCar (LiDAR 4096 points+RGB)
MinkLoc++ (LiDAR+RGB)
San Francisco Landmark Dataset
BoQ
Inside Out
KITTI
None
SVOX-Night
SVOX-Overcast
SVOX-Rain
SVOX-Snow
SVOX-Sun
VP Air
SegVLAD-PreT (M)