HyperAI超神経

プラットフォーム

Image Retrieval On Oxf105K

評価指標

MAP

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

		Paper Title
Offline Diffusion	95.2%	Efficient Image Retrieval via Decoupling Diffusion into Online and Offline Processing
DELF+FT+ATT+DIR+QE	88.5%	Large-Scale Image Retrieval with Attentive Deep Local Features
DIR+QE*	87.8%	Deep Image Retrieval: Learning global representations for image search
CNN+IME layer	87.2%	Iterative Manifold Embedding Layer Learned by Incomplete Data for Large-scale Image Retrieval
DELF+FT+ATT	82.6%	Large-Scale Image Retrieval with Attentive Deep Local Features
siaMAC+QE*	77.9%	CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples
R-MAC+R+QE	73.2%	Particular object retrieval with integral max-pooling of CNN activations
R-MAC	61.6%	Particular object retrieval with integral max-pooling of CNN activations
SIFT+IME layer	31.3%	Iterative Manifold Embedding Layer Learned by Incomplete Data for Large-scale Image Retrieval

0 of 9 row(s) selected.

Image Retrieval On Oxf105K | SOTA | HyperAI超神経