# Referring Image Segmentation on G-Ref UMD (test)

Source: [Referring Image Segmentation on G-Ref UMD (test)](<https://www.sota2.com/research/sota/referring-image-segmentation-on-g-ref-umd-test>)

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

Data updated: 2026-02-27T07:44:53.246Z

Evaluation of Referring Image Segmentation performance on the test split of the G-Ref UMD dataset, typically measured using Intersection over Union (IoU).

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 | Referring Image Segmentation |
| Dataset (including any stated split) | G-Ref UMD (test) |
| Metric | IoU |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 19 of 19 |
| 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

- [IoU](<https://www.sota2.com/research/sota/referring-image-segmentation-on-g-ref-umd-test.md?metric=IoU>): Higher is better; unit: PERCENT.

- [Precision @ 0.5 Threshold](<https://www.sota2.com/research/sota/referring-image-segmentation-on-g-ref-umd-test.md?metric=Precision%20%40%200.5%20Threshold>): Higher is better; unit: PERCENT.

## Recorded results

| Method | IoU | Result date | Report paper |
| --- | --- | --- | --- |
| VLT | 66.22 | Not recorded | 68fa6e091e209131cea3d96b |
| MaIL | 62.87 | Not recorded | 68fa6b4f1e209131cea2d89e |
| MaIL | 62.87 | Not recorded | 68fa6e091e209131cea3d96b |
| LAVT | 62.09 | Not recorded | 68fa6e091e209131cea3d96b |
| MaIL | 61.39 | Not recorded | 68fa6b4f1e209131cea2d89e |
| CRIS | 60.36 | Not recorded | 68fa6e091e209131cea3d96b |
| VLT | 57.73 | Not recorded | 68fa6e091e209131cea3d96b |
| VLT | 56.65 | Not recorded | 68fa6b4f1e209131cea2d89e |
| LTS | 54.25 | Not recorded | 68fa6b4f1e209131cea2d89e |
| LTS | 54.25 | Not recorded | 68fa6e091e209131cea3d96b |
| ISFP | 53 | Not recorded | 68fa6e091e209131cea3d96b |
| CGAN | 51.69 | Not recorded | 68fa6b4f1e209131cea2d89e |
| CGAN | 51.69 | Not recorded | 68fa6e091e209131cea3d96b |
| MCN | 49.4 | Not recorded | 68fa6b4f1e209131cea2d89e |
| MCN | 49.4 | Not recorded | 68fa6e091e209131cea3d96b |
| MAttNet | 48.61 | Not recorded | 68fa6b4f1e209131cea2d89e |
| MAttNet | 48.61 | Not recorded | 68fa6e091e209131cea3d96b |
| CAC | 46.95 | Not recorded | 68fa6b4f1e209131cea2d89e |
| CAC | 46.95 | Not recorded | 68fa6e091e209131cea3d96b |

### VLT

IoU: 66.22. Result ID: 69a14b6b35dfcabb00e613a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

Recorded evaluation configuration: {"Visual Backbone":"Swin-B","Textual Encoder":"BERT"}. This record does not assert independent reproduction.

### MaIL

IoU: 62.87. Result ID: 698e7dc6ffc3901de09e0a9f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

Recorded evaluation configuration: {"Resolution":"Fixed aspect ratio"}. This record does not assert independent reproduction.

### MaIL

IoU: 62.87. Result ID: 69a14b6b35dfcabb00e613a5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

Recorded evaluation configuration: {"Visual Backbone":"ViLT","Textual Encoder":"BERT","large-scale pre-training":"true"}. This record does not assert independent reproduction.

### LAVT

IoU: 62.09. Result ID: 69a14b6b35dfcabb00e613a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c2c1e209131cea328ba.md>)

[Implementation](<https://www.sota2.com/LAVT-RIS>)

Recorded evaluation configuration: {"Visual Backbone":"Swin-B","Textual Encoder":"BERT"}. This record does not assert independent reproduction.

### MaIL

IoU: 61.39. Result ID: 698e7dc6ffc3901de09e0a9e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

Recorded evaluation configuration: {"Resolution":"416x416"}. This record does not assert independent reproduction.

### CRIS

IoU: 60.36. Result ID: 69a14b6b35dfcabb00e613a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c011e209131cea319eb.md>)

Recorded evaluation configuration: {"Visual Backbone":"CLIP-R101","Textual Encoder":"CLIP","large-scale pre-training":"true"}. This record does not assert independent reproduction.

### VLT

IoU: 57.73. Result ID: 69a14b6b35dfcabb00e613a4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

Recorded evaluation configuration: {"Visual Backbone":"Darknet53","Textual Encoder":"bi-GRU"}. This record does not assert independent reproduction.

### VLT

IoU: 56.65. Result ID: 698e7dc6ffc3901de09e0a9d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a911e209131cea28f0a.md>)

[Implementation](<https://github.com/henghuiding/Vision-Language-Transformer>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LTS

IoU: 54.25. Result ID: 698e7dc6ffc3901de09e0a9c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa699a1e209131cea22fd0.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LTS

IoU: 54.25. Result ID: 69a14b6b35dfcabb00e613a3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa699a1e209131cea22fd0.md>)

Recorded evaluation configuration: {"Visual Backbone":"Darknet53","Textual Encoder":"bi-GRU"}. This record does not assert independent reproduction.

### ISFP

IoU: 53. Result ID: 69a14b6b35dfcabb00e613a2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c671e209131cea33e0b.md>)

Recorded evaluation configuration: {"Visual Backbone":"Darknet53","Textual Encoder":"Bi-GRU"}. This record does not assert independent reproduction.

### CGAN

IoU: 51.69. Result ID: 698e7dc6ffc3901de09e0a9b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CGAN

IoU: 51.69. Result ID: 69a14b6b35dfcabb00e613a1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

Recorded evaluation configuration: {"Visual Backbone":"DeepLab-R101","Textual Encoder":"bi-GRU"}. This record does not assert independent reproduction.

### MCN

IoU: 49.4. Result ID: 698e7dc6ffc3901de09e0a9a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d7e.md>)

[Implementation](<https://github.com/luogen1996/MCN>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MCN

IoU: 49.4. Result ID: 69a14b6b35dfcabb00e613a0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67361e209131cea14d7e.md>)

[Implementation](<https://github.com/luogen1996/MCN>)

Recorded evaluation configuration: {"Visual Backbone":"Darknet53","Textual Encoder":"bi-GRU"}. This record does not assert independent reproduction.

### MAttNet

IoU: 48.61. Result ID: 698e7dc6ffc3901de09e0a98.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642c1e209131cea031d4.md>)

[Implementation](<https://github.com/lichengunc/MAttNet>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MAttNet

IoU: 48.61. Result ID: 69a14b6b35dfcabb00e6139e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642c1e209131cea031d4.md>)

[Implementation](<https://github.com/lichengunc/MAttNet>)

Recorded evaluation configuration: {"Visual Backbone":"MaskRCNN-R101","Textual Encoder":"bi-LSTM"}. This record does not assert independent reproduction.

### CAC

IoU: 46.95. Result ID: 698e7dc6ffc3901de09e0a99.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6b4f1e209131cea2d89e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66801e209131cea10b04.md>)

[Implementation](<https://github.com/wenz116/lang2seg>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CAC

IoU: 46.95. Result ID: 69a14b6b35dfcabb00e6139f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d96b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66801e209131cea10b04.md>)

[Implementation](<https://github.com/wenz116/lang2seg>)

Recorded evaluation configuration: {"Visual Backbone":"ResNet101","Textual Encoder":"bi-LSTM"}. This record does not assert independent reproduction.

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