# Few-shot classification on Cars

Source: [Few-shot classification on Cars](<https://www.sota2.com/research/sota/few-shot-classification-on-cars>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:28:44.297Z

Evaluation of Few-shot Classification performance on the Cars dataset, reporting accuracy.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Few-shot classification |
| Dataset (including any stated split) | Cars |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 41 of 41 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/few-shot-classification-on-cars.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [1-shot Accuracy](<https://www.sota2.com/research/sota/few-shot-classification-on-cars.md?metric=1-shot%20Accuracy>): Higher is better; unit: PERCENT.

- [5-shot Accuracy](<https://www.sota2.com/research/sota/few-shot-classification-on-cars.md?metric=5-shot%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Supervised | 75.15 | Not recorded | 68fa6eaa1e209131cea41376 |
| PsCo | 57.5 | Not recorded | 68fa6eaa1e209131cea41376 |
| PsCo | 56.19 | Not recorded | 68fa6eaa1e209131cea41376 |
| PsCo | 53.45 | Not recorded | 68fa6eaa1e209131cea41376 |
| MoCo v2 | 51.77 | Not recorded | 68fa6eaa1e209131cea41376 |
| PsCo | 50.73 | Not recorded | 68fa6eaa1e209131cea41376 |
| SimCLR | 50.05 | Not recorded | 68fa6eaa1e209131cea41376 |
| SWAV | 50.03 | Not recorded | 68fa6eaa1e209131cea41376 |
| MAML | 48.82 | Not recorded | 68fa6eaa1e209131cea41376 |
| ProtoNets | 47.98 | Not recorded | 68fa6eaa1e209131cea41376 |
| MoCo v2 | 47.67 | Not recorded | 68fa6eaa1e209131cea41376 |
| MoCo v2 | 46.42 | Not recorded | 68fa6eaa1e209131cea41376 |
| Meta-SVEBM | 46.23 | Not recorded | 68fa6eaa1e209131cea41376 |
| MetaOptNet-SVM-trainval | 45.9 | Not recorded | 68fa67391e209131cea1507c |
| BYOL | 45.74 | Not recorded | 68fa6eaa1e209131cea41376 |
| GNN (FT) | 44.9 | Not recorded | 68fa67391e209131cea1507c |
| GNN | 44.28 | Not recorded | 68fa67391e209131cea1507c |
| PsCo | 44.01 | Not recorded | 68fa6eaa1e209131cea41376 |
| MatchingNet (FT) | 41.24 | Not recorded | 68fa67391e209131cea1507c |
| MAML | 41.17 | Not recorded | 68fa6eaa1e209131cea41376 |
| RelationNet (FT) | 39.91 | Not recorded | 68fa67391e209131cea1507c |
| MatchingNet | 38.99 | Not recorded | 68fa67391e209131cea1507c |
| ProtoNets | 38.98 | Not recorded | 68fa6eaa1e209131cea41376 |
| MoCo v2 | 38.65 | Not recorded | 68fa6eaa1e209131cea41376 |
| Meta-GMVAE | 38.36 | Not recorded | 68fa6eaa1e209131cea41376 |
| SimCLR | 37.4 | Not recorded | 68fa6eaa1e209131cea41376 |
| RelationNet | 37.33 | Not recorded | 68fa67391e209131cea1507c |
| SWAV | 36.85 | Not recorded | 68fa6eaa1e209131cea41376 |
| Meta-SVEBM | 34.27 | Not recorded | 68fa6eaa1e209131cea41376 |
| Meta-GMVAE | 31.39 | Not recorded | 68fa6eaa1e209131cea41376 |

### Supervised

Accuracy: 75.15. Result ID: 699038e83a6c314849ff02b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### PsCo

Accuracy: 57.5. Result ID: 699038e83a6c314849ff03a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Implementation](<https://github.com/alinlab/PsCo>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### PsCo

Accuracy: 56.19. Result ID: 699038e83a6c314849ff02b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Implementation](<https://github.com/alinlab/PsCo>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Base Method":"BYOL","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### PsCo

Accuracy: 53.45. Result ID: 699038e83a6c314849ff02af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Implementation](<https://github.com/alinlab/PsCo>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Base Method":"MoCo v2","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### MoCo v2

Accuracy: 51.77. Result ID: 699038e83a6c314849ff039c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### PsCo

Accuracy: 50.73. Result ID: 699038e83a6c314849ff02ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Implementation](<https://github.com/alinlab/PsCo>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Base Method":"MoCo v2","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### SimCLR

Accuracy: 50.05. Result ID: 699038e83a6c314849ff039a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d71e209131cea18de5.md>)

[Implementation](<https://github.com/google-research/simclr>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### SWAV

Accuracy: 50.03. Result ID: 699038e83a6c314849ff039e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa690b1e209131cea1faec.md>)

[Implementation](<https://github.com/facebookresearch/swav>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### MAML

Accuracy: 48.82. Result ID: 699038e83a6c314849ff03a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63861e209131ce9ff3e8.md>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### ProtoNets

Accuracy: 47.98. Result ID: 699038e83a6c314849ff03a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa636e1e209131ce9febd3.md>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### MoCo v2

Accuracy: 47.67. Result ID: 699038e83a6c314849ff02ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### MoCo v2

Accuracy: 46.42. Result ID: 699038e83a6c314849ff02ac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-18","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### Meta-SVEBM

Accuracy: 46.23. Result ID: 699038e83a6c314849ff0398.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### MetaOptNet-SVM-trainval

Accuracy: 45.9. Result ID: 698d22fb8b517eb4331f4749.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65a91e209131cea0bba9.md>)

[Implementation](<https://github.com/kjunelee/MetaOptNet>)

Recorded evaluation configuration: {"Backbone":"ResNet-12","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":false}. This record does not assert independent reproduction.

### BYOL

Accuracy: 45.74. Result ID: 699038e83a6c314849ff02b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa684a1e209131cea1b1e9.md>)

[Implementation](<https://github.com/deepmind/deepmind-research/tree/byol>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Evaluation protocol":"5-way 5-shot"}. This record does not assert independent reproduction.

### GNN (FT)

Accuracy: 44.9. Result ID: 698d22fb8b517eb4331f474f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Implementation](<https://github.com/hytseng0509/CrossDomainFewShot>)

Recorded evaluation configuration: {"Backbone":"ResNet-10","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":true}. This record does not assert independent reproduction.

### GNN

Accuracy: 44.28. Result ID: 698d22fb8b517eb4331f474e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Implementation](<https://github.com/hytseng0509/CrossDomainFewShot>)

Recorded evaluation configuration: {"Backbone":"ResNet-10","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":false}. This record does not assert independent reproduction.

### PsCo

Accuracy: 44.01. Result ID: 699038e83a6c314849ff039f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Implementation](<https://github.com/alinlab/PsCo>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### MatchingNet (FT)

Accuracy: 41.24. Result ID: 698d22fb8b517eb4331f474b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Implementation](<https://github.com/hytseng0509/CrossDomainFewShot>)

Recorded evaluation configuration: {"Backbone":"ResNet-10","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":true}. This record does not assert independent reproduction.

### MAML

Accuracy: 41.17. Result ID: 699038e83a6c314849ff03a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63861e209131ce9ff3e8.md>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### RelationNet (FT)

Accuracy: 39.91. Result ID: 698d22fb8b517eb4331f474d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Implementation](<https://github.com/hytseng0509/CrossDomainFewShot>)

Recorded evaluation configuration: {"Backbone":"ResNet-10","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":true}. This record does not assert independent reproduction.

### MatchingNet

Accuracy: 38.99. Result ID: 698d22fb8b517eb4331f474a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Implementation](<https://github.com/hytseng0509/CrossDomainFewShot>)

Recorded evaluation configuration: {"Backbone":"ResNet-10","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":false}. This record does not assert independent reproduction.

### ProtoNets

Accuracy: 38.98. Result ID: 699038e83a6c314849ff03a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa636e1e209131ce9febd3.md>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### MoCo v2

Accuracy: 38.65. Result ID: 699038e83a6c314849ff039b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### Meta-GMVAE

Accuracy: 38.36. Result ID: 699038e83a6c314849ff0396.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"way":5,"shot":20,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### SimCLR

Accuracy: 37.4. Result ID: 699038e83a6c314849ff0399.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67d71e209131cea18de5.md>)

[Implementation](<https://github.com/google-research/simclr>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### RelationNet

Accuracy: 37.33. Result ID: 698d22fb8b517eb4331f474c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67391e209131cea1507c.md>)

[Implementation](<https://github.com/hytseng0509/CrossDomainFewShot>)

Recorded evaluation configuration: {"Backbone":"ResNet-10","Training Dataset":"mini-ImageNet","FT (Feature-wise Transformation)":false}. This record does not assert independent reproduction.

### SWAV

Accuracy: 36.85. Result ID: 699038e83a6c314849ff039d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa690b1e209131cea1faec.md>)

[Implementation](<https://github.com/facebookresearch/swav>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### Meta-SVEBM

Accuracy: 34.27. Result ID: 699038e83a6c314849ff0397.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

### Meta-GMVAE

Accuracy: 31.39. Result ID: 699038e83a6c314849ff0395.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6eaa1e209131cea41376.md>)

Recorded evaluation configuration: {"way":5,"shot":5,"Backbone":"Conv5","Pre-trained":"miniImageNet"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.