# Amodal 3D Object Proposal on NYU v2 (test)

Source: [Amodal 3D Object Proposal on NYU v2 (test)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test>)

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

Data updated: 2026-02-26T06:01:18.950Z

Evaluation of Amodal 3D Object Proposal performance on the test split of the NYU Depth V2 dataset, typically measured using class-specific Recall metrics.

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 | Amodal 3D Object Proposal |
| Dataset (including any stated split) | NYU v2 (test) |
| Metric | Recall (sofa) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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

- [Recall (sofa)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(sofa)>): Higher is better; unit: PERCENT.

- [Recall (bed)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(bed)>): Higher is better; unit: PERCENT.

- [Recall (bathtub)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(bathtub)>): Higher is better; unit: PERCENT.

- [Recall (garbage bin)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(garbage%20bin)>): Higher is better; unit: PERCENT.

- [Recall (chair)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(chair)>): Higher is better; unit: PERCENT.

- [Recall (desk)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(desk)>): Higher is better; unit: PERCENT.

- [Recall (pillow)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(pillow)>): Higher is better; unit: PERCENT.

- [Recall (bookshelf)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(bookshelf)>): Higher is better; unit: PERCENT.

- [Recall (table)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(table)>): Higher is better; unit: PERCENT.

- [Recall (box)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(box)>): Higher is better; unit: PERCENT.

- [Recall (monitor)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(monitor)>): Higher is better; unit: PERCENT.

- [Recall (night stand)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(night%20stand)>): Higher is better; unit: PERCENT.

- [Recall (door)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(door)>): Higher is better; unit: PERCENT.

- [Recall (lamp)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(lamp)>): Higher is better; unit: PERCENT.

- [Recall (sink)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(sink)>): Higher is better; unit: PERCENT.

- [Recall (toilet)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(toilet)>): Higher is better; unit: PERCENT.

- [Recall (dresser)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(dresser)>): Higher is better; unit: PERCENT.

- [Recall (cabinet)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(cabinet)>): Higher is better; unit: PERCENT.

- [Recall (tv)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Recall%20(tv)>): Higher is better; unit: PERCENT.

- [AR (Average Recall)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=AR%20(Average%20Recall)>): Higher is better; unit: PERCENT.

- [ABO (Amodal Bounding Box Overlap)](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=ABO%20(Amodal%20Bounding%20Box%20Overlap)>): Higher is better; unit: PERCENT.

- [Num Boxes](<https://www.sota2.com/research/sota/amodal-3d-object-proposal-on-nyu-v2-test.md?metric=Num%20Boxes>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Recall (sofa) | Result date | Report paper |
| --- | --- | --- | --- |
| 3D RPN | 100 | Not recorded | 68fa62611e209131ce9f8f21 |
| 3D RPN | 100 | Not recorded | 68fa62611e209131ce9f8f21 |
| All Anchors | 100 | Not recorded | 68fa62611e209131ce9f8f21 |
| 3D RPN | 87.5 | Not recorded | 68fa62611e209131ce9f8f21 |
| 3D Selective Search | 79.2 | Not recorded | 68fa62611e209131ce9f8f21 |
| 2D To 3D | 41.7 | Not recorded | 68fa62611e209131ce9f8f21 |

### 3D RPN

Recall (sofa): 100. Result ID: 698bd70d7a967eddb771b930.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

[Implementation](<http://dss.cs.princeton.edu/>)

Recorded evaluation configuration: {"IoU Threshold":0.25,"scale":"multi"}. This record does not assert independent reproduction.

### 3D RPN

Recall (sofa): 100. Result ID: 698bd70d7a967eddb771b931.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

[Implementation](<http://dss.cs.princeton.edu/>)

Recorded evaluation configuration: {"IoU Threshold":0.25,"scale":"multi","input":"RGB + TSDF"}. This record does not assert independent reproduction.

### All Anchors

Recall (sofa): 100. Result ID: 698bd70d7a967eddb771b932.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

Recorded evaluation configuration: {"IoU Threshold":0.25,"mode":"Performance Upper Bound"}. This record does not assert independent reproduction.

### 3D RPN

Recall (sofa): 87.5. Result ID: 698bd70d7a967eddb771b92f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

[Implementation](<http://dss.cs.princeton.edu/>)

Recorded evaluation configuration: {"IoU Threshold":0.25,"scale":"single"}. This record does not assert independent reproduction.

### 3D Selective Search

Recall (sofa): 79.2. Result ID: 698bd70d7a967eddb771b92e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

Recorded evaluation configuration: {"IoU Threshold":0.25}. This record does not assert independent reproduction.

### 2D To 3D

Recall (sofa): 41.7. Result ID: 698bd70d7a967eddb771b92d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62611e209131ce9f8f21.md>)

Recorded evaluation configuration: {"IoU Threshold":0.25,"source\_proposals":"2D proposals from \[9\]"}. This record does not assert independent reproduction.

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