# 3D Object Detection on ScanRefer (val)

Source: [3D Object Detection on ScanRefer (val)](<https://www.sota2.com/research/sota/3d-object-detection-on-scanrefer-val>)

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

Data updated: 2026-07-08T07:34:49.384Z

Evaluation of 3D Object Detection performance on the validation split of the ScanRefer dataset, reporting metrics like mAP and AR at IoU thresholds of 0.25 and 0.5.

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 | 3D Object Detection |
| Dataset (including any stated split) | ScanRefer (val) |
| Metric | mAP @ IoU=0.25 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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 @ IoU=0.25](<https://www.sota2.com/research/sota/3d-object-detection-on-scanrefer-val.md?metric=mAP%20%40%20IoU%3D0.25>): Higher is better; unit: PERCENT.

- [AR @ IoU=0.25](<https://www.sota2.com/research/sota/3d-object-detection-on-scanrefer-val.md?metric=AR%20%40%20IoU%3D0.25>): Higher is better; unit: PERCENT.

- [mAP @ IoU=0.5](<https://www.sota2.com/research/sota/3d-object-detection-on-scanrefer-val.md?metric=mAP%20%40%20IoU%3D0.5>): Higher is better; unit: PERCENT.

- [AR @ IoU=0.5](<https://www.sota2.com/research/sota/3d-object-detection-on-scanrefer-val.md?metric=AR%20%40%20IoU%3D0.5>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mAP @ IoU=0.25 | Result date | Report paper |
| --- | --- | --- | --- |
| PVCap-Swin3D | 75.65 | Not recorded | 6a4dd98e4da6617fe9198b26 |
| PVCap-SparseUNet | 71.75 | Not recorded | 6a4dd98e4da6617fe9198b26 |
| Vote2Cap++ | 70.52 | Not recorded | 6a4dd98e4da6617fe9198b26 |
| Vote2Cap | 69.61 | Not recorded | 6a4dd98e4da6617fe9198b26 |

### PVCap-Swin3D

mAP @ IoU=0.25: 75.65. Result ID: 6a4dfd8960b88bb339d73b84.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd98e4da6617fe9198b26.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4dd98e4da6617fe9198b26.md>)

Recorded evaluation configuration: {"Backbone":"Swin3D","Additional 2D features":"false"}. This record does not assert independent reproduction.

### PVCap-SparseUNet

mAP @ IoU=0.25: 71.75. Result ID: 6a4dfd8960b88bb339d73b83.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd98e4da6617fe9198b26.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4dd98e4da6617fe9198b26.md>)

Recorded evaluation configuration: {"Backbone":"SparseUNet","Additional 2D features":"false"}. This record does not assert independent reproduction.

### Vote2Cap++

mAP @ IoU=0.25: 70.52. Result ID: 6a4dfd8960b88bb339d73b82.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd98e4da6617fe9198b26.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa707b1e209131cea4b5e2.md>)

[Implementation](<https://github.com/ch3cook-fdu/Vote2Cap-DETR>)

Recorded evaluation configuration: {"Additional 2D features":"false"}. This record does not assert independent reproduction.

### Vote2Cap

mAP @ IoU=0.25: 69.61. Result ID: 6a4dfd8960b88bb339d73b81.

[Reporting paper](<https://www.sota2.com/research/paper/6a4dd98e4da6617fe9198b26.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e5d1e209131cea3f5fd.md>)

[Implementation](<https://github.com/ch3cook-fdu/Vote2Cap-DETR>)

Recorded evaluation configuration: {"Additional 2D features":"false"}. This record does not assert independent reproduction.

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