Browse › Computer Vision › Metric Learning › CUB-200-2011
CUB-200-2011 Benchmark (Metric Learning)
The goal of Metric Learning is to learn a representation function that maps objects into an embedded space. The distance in the embedded space should preserve the objects’ similarity — similar objects get close and dissimilar objects get far away. Various loss functions have been developed for Metric Learning. For example, the contrastive loss guides the objects from the same class to be mapped to the same point and those from different classes to be mapped to different points whose distances are larger than a margin. Triplet loss is also popular, which requires the distance between the anchor sample and the positive sample to be smaller than the distance between the anchor sample and the negative sample.
Source: Road Network Metric Learning for Estimated Time of Arrival
The archive carries no text for this table; the description above is the archive's text for the task Metric Learning. archive 2025-07-28
Over time archive 2025-07-28
The chart needs JavaScript; the table below carries every value.
Not inferred: R@1. Points are placed at the row's paper date; 30 of 30 rows carry one.
Results archive 2025-07-28
Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.
| Paper | Code | Ran Syntology | Report | |||||
|---|---|---|---|---|---|---|---|---|
| 1 | Unicom+ViT-L@336px | 90.1 | ✓ | Paper | Code | 2023 | 3 of 6 ran · 3 unverified | report |
| 2 | EfficientDML-VPTSP-G/512 | 88.5 | ✓ | Paper | Code | 2024 | 3 of 5 ran · 2 unverified | report |
| 3 | Hyp-ViT | 85.6 | ✓ | Paper | Code | 2022 | 4 of 6 ran · 2 unverified | report |
| 4 | NED | 74.9 | ✓ | Paper | – | 2020 | no code linked | report |
| 5 | CCL (ResNet-50) | 73.45 | – | Paper | – | 2023 | no code linked | report |
| 6 | ResNet-50 + AVSL | 71.9 | ✓ | Paper | Code | 2022 | 0 of 4 ran · 4 unverified | report |
| 7 | ResNet-50 + Intra-Batch Connections | 71.8 | ✓ | Paper | Code | 2021 | linked, not harvested | report |
| 8 | ResNet-50 + Metrix | 71.4 | ✓ | Paper | Code | 2021 | 2 of 2 ran · 0 unverified | report |
| 9 | ResNet50 + Language | 71.4 | ✓ | Paper | Code | 2022 | 0 of 5 ran · 5 unverified | report |
| 10 | BN-Inception + Proxy-Anchor | 71.1 | ✓ | Paper | Code | 2020 | linked, not harvested | report |
| 11 | ResNet50 + NIR | 70.5 | ✓ | Paper | Code | 2022 | 0 of 3 ran · 3 unverified | report |
| 12 | ResNet50 + S2SD | 70.1 | ✓ | Paper | Code | 2020 | 0 of 3 ran · 3 unverified | report |
| 13 | ResNet-50 + Cross-Entropy | 69.2 | ✓ | Paper | Code | 2020 | 1 of 1 ran · 0 unverified | report |
| 14 | ResNet50 + DiVA | 69.2 | ✓ | Paper | Code | 2020 | 2 of 5 ran · 3 unverified | report |
| 15 | ResNet-50 + ProxyNCA++ | 69.0 | ✓ | Paper | Code | 2020 | 0 of 5 ran · 5 unverified | report |
| 16 | MS + DIML | 68.15 | – | Paper | Code | 2021 | 2 of 2 ran · 0 unverified | report |
| 17 | ResNet50 (128) + PADS | 67.3 | ✓ | Paper | Code | 2020 | 0 of 10 ran · 10 unverified | report |
| 18 | ABE + HORDE | 66.8 | ✓ | Paper | Code | 2019 | 0 of 8 ran · 8 unverified | report |
| 19 | ResNet50 (128) + MIC | 66.1 | ✓ | Paper | Code | 2019 | linked, not harvested | report |
| 20 | BN-Inception + Group Loss | 65.5 | ✓ | Paper | Code | 2019 | 3 of 18 ran · 15 unverified | report |
| 21 | BN-Inception + SoftTriple | 65.4 | ✓ | Paper | Code | 2019 | 2 of 4 ran · 2 unverified | report |
| 22 | EPSHN(512) | 64.9 | – | Paper | Code | 2019 | 1 of 1 ran · 0 unverified | report |
| 23 | Gradient Surgery | 63.8 | – | Paper | – | 2022 | no code linked | report |
| 24 | ResNet-50 + Margin | 63.6 | ✓ | Paper | Code | 2017 | 2 of 2 ran · 0 unverified | report |
| 25 | HDC | 60.7 | – | Paper | Code | 2016 | linked, not harvested | report |
| 26 | ABE-8-512 | 60.6 | ✓ | Paper | – | 2018 | no code linked | report |
| 27 | PDDM Quadruplet | 58.3 | – | Paper | – | 2016 | no code linked | report |
| 28 | SCT(64) | 57.7 | – | Paper | Code | 2020 | linked, not harvested | report |
| 29 | EPSHN(64) | 57.3 | – | Paper | Code | 2019 | 1 of 1 ran · 0 unverified | report |
| 30 | GoogLeNet + HDML | 53.7 | ✓ | Paper | Code | 2019 | 0 of 2 ran · 2 unverified | report |
All 30 rows shown. 30 link to a paper page on this site; 22 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28
Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 20 rows have a graph line, from 19 distinct papers; 12 rows (11 papers) have at least one sample that ran. Counting each paper once: Syntology ran 25 of 92 samples; 67 unverified. Separately, 14 of those 92 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections