List your company
ResearchFormerly WizWand

Place Recognition on Oxford Robotcar (test)

96.4Avg Recall @1%SOE-Net
44.524857.992471.4684.9276Dec 11, 2018Apr 9, 2019Aug 6, 2019Dec 3, 2019Mar 31, 2020Jul 28, 2020Nov 24, 2020
Updated 8mo ago

Evaluation results

MethodLinks
SOE-NetDescriptor Dimension=256-dim, Training Data=Oxford RobotCar2020.1196.4—
CORAL-VLADFusion strategy=element-wise concatenation in four residual blocks, Alternative Name=Con-Four2020.1196.1388.93
FN-SF-VLADaggregation_structure=SF, representing=LPD-Net2018.1294.9286.28
LPD-NetDescriptor Dimension=256-dim, Training Data=Oxford RobotCar2020.1194.92—
Sum-FourFusion strategy=element-wise summarization in four residual blocks2020.1194.4386.23
LPD-Net2020.1194.4286.28
Con-FirstFusion strategy=element-wise concatenation in the first residual block2020.1193.6284.82
Ele-VLADStream=Elevation image NetVLAD2020.1193.6182.49
Sum-FirstFusion strategy=element-wise summarization in the first residual block2020.1192.7182.44
FN-PC-VLADaggregation_structure=PC2018.1292.2781.41
FN-DG-VLADgraph_type=Dynamic Graph2018.1291.4480.14
Aug-Net2020.1191.2479.47
FN-PM-VLADaggregation_structure=PM2018.1291.278.77
FN-NG-VLADgraph_type=Neighbor Graph2018.1290.3877.74
FN-VLADmodule=without graph-based neighborhood aggregation2018.1289.7775.79
DAGCDescriptor Dimension=256-dim, Training Data=Oxford RobotCar2020.1187.49—
Img-VLAD2020.1185.2464.47
DH3DDescriptor Dimension=256-dim, Training Data=Oxford RobotCar2020.1184.26—
PCANDescriptor Dimension=256-dim, Training Data=Oxford RobotCar2020.1183.81—
Vis-VLADStream=Vision NetVLAD2020.1183.0557.62
PN-VLAD baseline2018.1281.0162.76
PN-VLAD2020.1181.0167.94
PN-VLAD refine2018.1280.7163.33
PointNetVLADDescriptor Dimension=256-dim, Training Data=Oxford RobotCar2020.1180.31—
NN-VLADclustering=kNN (k=20)2018.1279.2161.96
PN MAX2018.1273.8754.16
PN STD2018.1246.5231.87