|
Unicom+ViT-L@336px
|
Unicom: Universal and Compact Representation Lear…
|
91.20
|
2023-04-12
|
|
|
STIR
|
STIR: Siamese Transformer for Image Retrieval Pos…
|
88.30
|
2023-04-26
|
|
|
Recall@k Surrogate Loss (ViT-B/16)
|
Recall@k Surrogate Loss with Large Batches and Si…
|
88.00
|
2021-08-25
|
|
|
ViT-Triplet
|
STIR: Siamese Transformer for Image Retrieval Pos…
|
86.50
|
2023-04-26
|
|
|
ROADMAP (DeiT-S)
|
Robust and Decomposable Average Precision for Ima…
|
86.00
|
2021-10-01
|
|
|
Hyp-ViT
|
Hyperbolic Vision Transformers: Combining Improve…
|
85.90
|
2022-03-21
|
|
|
Hyp-DINO
|
Hyperbolic Vision Transformers: Combining Improve…
|
85.10
|
2022-03-21
|
|
|
Recall@k Surrogate Loss (ViT-B/32)
|
Recall@k Surrogate Loss with Large Batches and Si…
|
85.10
|
2021-08-25
|
|
|
ROADMAP (ResNet-50)
|
Robust and Decomposable Average Precision for Ima…
|
83.10
|
2021-10-01
|
|
|
CCL (ResNet-50)
|
Center Contrastive Loss for Metric Learning
|
83.10
|
2023-08-01
|
|
|
Recall@k Surrogate Loss (ResNet-50)
|
Recall@k Surrogate Loss with Large Batches and Si…
|
82.70
|
2021-08-25
|
|
|
Gradient Surgery
|
Dissecting the impact of different loss functions…
|
82.30
|
2022-01-27
|
|
|
HAPPIER_F
|
Hierarchical Average Precision Training for Perti…
|
81.80
|
2022-07-05
|
|
|
SCT(512)
|
Hard negative examples are hard, but useful
|
81.60
|
2020-07-24
|
|
|
ResNet-50 + Metrix
|
It Takes Two to Tango: Mixup for Deep Metric Lear…
|
81.30
|
2021-06-09
|
|
|
ResNet50 + Language
|
Integrating Language Guidance into Vision-based D…
|
81.30
|
2022-03-16
|
|
|
NED
|
Calibrated neighborhood aware confidence measure …
|
81.20
|
2020-06-08
|
|
|
ResNet-50 + Cross-Entropy
|
A unifying mutual information view of metric lear…
|
81.10
|
2020-03-19
|
|
|
ResNet50 + S2SD
|
S2SD: Simultaneous Similarity-based Self-Distilla…
|
81.00
|
2020-09-17
|
|
|
HAPPIER
|
Hierarchical Average Precision Training for Perti…
|
81.00
|
2022-07-05
|
|
|
ResNet-50 + ProxyNCA++
|
ProxyNCA++: Revisiting and Revitalizing Proxy Nei…
|
80.70
|
2020-04-02
|
|
|
ResNet50 + NIR
|
Non-isotropy Regularization for Proxy-based Deep …
|
80.70
|
2022-03-16
|
|
|
MS + DAS (K=8)
|
DAS: Densely-Anchored Sampling for Deep Metric Le…
|
80.59
|
2022-07-30
|
|
|
BN-Inception + Proxy-Anchor
|
Proxy Anchor Loss for Deep Metric Learning
|
80.30
|
2020-03-31
|
|
|
ResNet50 + DiVA
|
DiVA: Diverse Visual Feature Aggregation for Deep…
|
79.60
|
2020-04-28
|
|
|
ResNet50 + AVSL
|
Attributable Visual Similarity Learning
|
79.60
|
2022-03-28
|
|
|
Margin + DIML
|
Towards Interpretable Deep Metric Learning with S…
|
79.26
|
2021-08-12
|
|
|
Circle Loss
|
Circle Loss: A Unified Perspective of Pair Simila…
|
78.30
|
2020-02-25
|
|
|
EPSHN(512)
|
Improved Embeddings with Easy Positive Triplet Mi…
|
78.30
|
2019-04-08
|
|
|
QB-Norm+RDML
|
Cross Modal Retrieval with Querybank Normalisation
|
78.10
|
2021-12-23
|
|
|
ResNet50 (128) + MIC
|
MIC: Mining Interclass Characteristics for Improv…
|
77.20
|
2019-09-25
|
|
|
ResNet50 (128) + PADS
|
PADS: Policy-Adapted Sampling for Visual Similari…
|
76.50
|
2020-03-24
|
|
|
Group Loss
|
The Group Loss for Deep Metric Learning
|
75.70
|
2019-12-01
|
|