# Image Retrieval on Revisited Paris (RParis) (test)

Source: [Image Retrieval on Revisited Paris (RParis) (test)](<https://www.sota2.com/research/sota/image-retrieval-on-revisited-paris-rparis-test>)

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

Data updated: 2026-02-26T06:09:51.978Z

Evaluation of Image Retrieval performance on the test split of the Revisited Paris (RParis) dataset, reporting mAP scores for Medium and Hard queries.

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 | Image Retrieval |
| Dataset (including any stated split) | Revisited Paris (RParis) (test) |
| Metric | mAP (Medium) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [mAP (Medium)](<https://www.sota2.com/research/sota/image-retrieval-on-revisited-paris-rparis-test.md?metric=mAP%20(Medium)>): Higher is better; unit: PERCENT.

- [mAP (Hard)](<https://www.sota2.com/research/sota/image-retrieval-on-revisited-paris-rparis-test.md?metric=mAP%20(Hard)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mAP (Medium) | Result date | Report paper |
| --- | --- | --- | --- |
| ResNet-101 | 69.4 | Not recorded | 68fa6e7d1e209131cea4042f |
| SAIG-D | 68.2 | Not recorded | 68fa6e7d1e209131cea4042f |
| ResNet-101 | 66 | Not recorded | 68fa6e7d1e209131cea4042f |
| SAIG-D | 65.9 | Not recorded | 68fa6e7d1e209131cea4042f |
| ResNet-101 | 63.4 | Not recorded | 68fa6e7d1e209131cea4042f |
| ResNet-101 | 62.6 | Not recorded | 68fa6e7d1e209131cea4042f |
| ResNet-50 | 61.2 | Not recorded | 68fa6e7d1e209131cea4042f |
| ResNet-50 | 60.6 | Not recorded | 68fa6e7d1e209131cea4042f |

### ResNet-101

mAP (Medium): 69.4. Result ID: 699032f23a6c314849fed818.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Input Size":"384 x 384","Backbone":"R101","Pooling Layer":"GeM","Dim":"2048"}. This record does not assert independent reproduction.

### SAIG-D

mAP (Medium): 68.2. Result ID: 699032f23a6c314849fed81a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Implementation](<https://github.com/yanghongji2007/SAIG>)

Recorded evaluation configuration: {"Input Size":"384 x 384","Backbone":"SAIG-D","Pooling Layer":"GAP","Dim":"384"}. This record does not assert independent reproduction.

### ResNet-101

mAP (Medium): 66. Result ID: 699032f23a6c314849fed819.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Input Size":"384 x 384","Backbone":"R101","Pooling Layer":"R-MAC","Dim":"2048"}. This record does not assert independent reproduction.

### SAIG-D

mAP (Medium): 65.9. Result ID: 699032f23a6c314849fed817.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Implementation](<https://github.com/yanghongji2007/SAIG>)

Recorded evaluation configuration: {"Input Size":"224 x 224","Backbone":"SAIG-D","Pooling Layer":"GAP","Dim":"384"}. This record does not assert independent reproduction.

### ResNet-101

mAP (Medium): 63.4. Result ID: 699032f23a6c314849fed815.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Input Size":"224 x 224","Backbone":"R101","Pooling Layer":"GeM","Dim":"2048"}. This record does not assert independent reproduction.

### ResNet-101

mAP (Medium): 62.6. Result ID: 699032f23a6c314849fed816.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Input Size":"224 x 224","Backbone":"R101","Pooling Layer":"R-MAC","Dim":"2048"}. This record does not assert independent reproduction.

### ResNet-50

mAP (Medium): 61.2. Result ID: 699032f23a6c314849fed813.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Input Size":"224 x 224","Backbone":"R50","Pooling Layer":"GeM","Dim":"2048"}. This record does not assert independent reproduction.

### ResNet-50

mAP (Medium): 60.6. Result ID: 699032f23a6c314849fed814.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e7d1e209131cea4042f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62321e209131ce9f8328.md>)

Recorded evaluation configuration: {"Input Size":"224 x 224","Backbone":"R50","Pooling Layer":"R-MAC","Dim":"2048"}. This record does not assert independent reproduction.

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