SOTAVerified

GPS Embeddings

GPS Embeddings is the collective name for a set of feature-learning techniques where GPS coordinates are mapped to vectors of real numbers.

Papers

No papers found.

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1GeoCLIPmAP0.36Unverified
2GPS2Vec+mAP0.35Unverified
3Liao et al.mAP0.35Unverified
4GPS2VecmAP0.3Unverified
5GPS2Vec onehotmAP0.28Unverified
6HashTagmAP0.26Unverified
7TagFeaturemAP0.26Unverified
8OneHotmAP0.24Unverified
#ModelMetricClaimedVerifiedStatus
1GeoCLIP mAP0.25Unverified
2GPS2Vec+ mAP0.21Unverified
3GPS2Vec mAP0.18Unverified
4HashTag mAP0.16Unverified
5TagFeature mAP0.16Unverified
6GPS2Vec onehot mAP0.13Unverified
7OneHot mAP0.07Unverified