Visual Place Recognition On Nardo Air

Recall@1

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

		Paper Title
AnyLoc-VLAD-DINOv2	76.06	AnyLoc: Towards Universal Visual Place Recognition
DINOv2	73.24	DINOv2: Learning Robust Visual Features without Supervision
DINO	57.75	Emerging Properties in Self-Supervised Vision Transformers
CLIP	42.25	AnyLoc: Towards Universal Visual Place Recognition
MixVPR	32.39	MixVPR: Feature Mixing for Visual Place Recognition
NetVLAD	19.72	NetVLAD: CNN architecture for weakly supervised place recognition
CosPlace	0	Rethinking Visual Geo-localization for Large-Scale Applications

0 of 7 row(s) selected.