# Fine-grained Semantic Segmentation on PartNet 1.0 (test)

Source: [Fine-grained Semantic Segmentation on PartNet 1.0 (test)](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test>)

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

Data updated: 2026-02-26T06:04:19.527Z

Evaluation of Fine-grained Semantic Segmentation performance on the test split of the PartNet 1.0 dataset, reporting shape mean Intersection over Union (mIoU) across various object parts.

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 | Fine-grained Semantic Segmentation |
| Dataset (including any stated split) | PartNet 1.0 (test) |
| Metric | Shape mIoU: Avg |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 12 of 12 |
| 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

- [Shape mIoU: Avg](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Avg>): Higher is better; unit: PERCENT.

- [Shape mIoU: Bag](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Bag>): Higher is better; unit: PERCENT.

- [Shape mIoU: Bed](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Bed>): Higher is better; unit: PERCENT.

- [Shape mIoU: Bott](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Bott>): Higher is better; unit: PERCENT.

- [Shape mIoU: Bowl](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Bowl>): Higher is better; unit: PERCENT.

- [Shape mIoU: Chair](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Chair>): Higher is better; unit: PERCENT.

- [Shape mIoU: Clock](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Clock>): Higher is better; unit: PERCENT.

- [Shape mIoU: Dish](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Dish>): Higher is better; unit: PERCENT.

- [Shape mIoU: Disp](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Disp>): Higher is better; unit: PERCENT.

- [Shape mIoU: Door](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Door>): Higher is better; unit: PERCENT.

- [Shape mIoU: Ear](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Ear>): Higher is better; unit: PERCENT.

- [Shape mIoU: Fauc](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Fauc>): Higher is better; unit: PERCENT.

- [Shape mIoU: Hat](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Hat>): Higher is better; unit: PERCENT.

- [Shape mIoU: Key](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Key>): Higher is better; unit: PERCENT.

- [Shape mIoU: Knife](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Knife>): Higher is better; unit: PERCENT.

- [Shape mIoU: Lamp](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Lamp>): Higher is better; unit: PERCENT.

- [Shape mIoU: Lap](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Lap>): Higher is better; unit: PERCENT.

- [Shape mIoU: Micro](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Micro>): Higher is better; unit: PERCENT.

- [Shape mIoU: Mug](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Mug>): Higher is better; unit: PERCENT.

- [Shape mIoU: Frid](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Frid>): Higher is better; unit: PERCENT.

- [Shape mIoU: Scis](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Scis>): Higher is better; unit: PERCENT.

- [Shape mIoU: Stora](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Stora>): Higher is better; unit: PERCENT.

- [Shape mIoU: Table](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Table>): Higher is better; unit: PERCENT.

- [Shape mIoU: Trash](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Trash>): Higher is better; unit: PERCENT.

- [Shape mIoU: Vase](<https://www.sota2.com/research/sota/fine-grained-semantic-segmentation-on-partnet-1-0-test.md?metric=Shape%20mIoU%3A%20Vase>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Shape mIoU: Avg | Result date | Report paper |
| --- | --- | --- | --- |
| PointNet++ | 76.8 | Not recorded | 68fa65301e209131cea08bc3 |
| PointCNN | 75.5 | Not recorded | 68fa65301e209131cea08bc3 |
| SpiderCNN | 73.9 | Not recorded | 68fa65301e209131cea08bc3 |
| PointNet | 71.8 | Not recorded | 68fa65301e209131cea08bc3 |
| PointNet++ | 54.7 | Not recorded | 68fa65301e209131cea08bc3 |
| PointNet++ | 53.4 | Not recorded | 68fa65301e209131cea08bc3 |
| SpiderCNN | 53.3 | Not recorded | 68fa65301e209131cea08bc3 |
| PointCNN | 52.1 | Not recorded | 68fa65301e209131cea08bc3 |
| PointNet | 50.1 | Not recorded | 68fa65301e209131cea08bc3 |
| PointCNN | 49.6 | Not recorded | 68fa65301e209131cea08bc3 |
| PointNet | 48.2 | Not recorded | 68fa65301e209131cea08bc3 |
| SpiderCNN | 48 | Not recorded | 68fa65301e209131cea08bc3 |

### PointNet++

Shape mIoU: Avg: 76.8. Result ID: 698ca09c0b614d12cff20f9e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63641e209131ce9fe5a5.md>)

Recorded evaluation configuration: {"Segmentation Level":"coarse (1)"}. This record does not assert independent reproduction.

### PointCNN

Shape mIoU: Avg: 75.5. Result ID: 698ca09c0b614d12cff20fa4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

Recorded evaluation configuration: {"Segmentation Level":"coarse (1)"}. This record does not assert independent reproduction.

### SpiderCNN

Shape mIoU: Avg: 73.9. Result ID: 698ca09c0b614d12cff20fa1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64d91e209131cea06a9f.md>)

Recorded evaluation configuration: {"Segmentation Level":"coarse (1)"}. This record does not assert independent reproduction.

### PointNet

Shape mIoU: Avg: 71.8. Result ID: 698ca09c0b614d12cff20f9b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634f1e209131ce9fe019.md>)

Recorded evaluation configuration: {"Segmentation Level":"coarse (1)"}. This record does not assert independent reproduction.

### PointNet++

Shape mIoU: Avg: 54.7. Result ID: 698ca09c0b614d12cff20f9f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63641e209131ce9fe5a5.md>)

Recorded evaluation configuration: {"Segmentation Level":"middle (2)"}. This record does not assert independent reproduction.

### PointNet++

Shape mIoU: Avg: 53.4. Result ID: 698ca09c0b614d12cff20fa0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63641e209131ce9fe5a5.md>)

Recorded evaluation configuration: {"Segmentation Level":"fine-grained (3)"}. This record does not assert independent reproduction.

### SpiderCNN

Shape mIoU: Avg: 53.3. Result ID: 698ca09c0b614d12cff20fa2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64d91e209131cea06a9f.md>)

Recorded evaluation configuration: {"Segmentation Level":"middle (2)"}. This record does not assert independent reproduction.

### PointCNN

Shape mIoU: Avg: 52.1. Result ID: 698ca09c0b614d12cff20fa5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

Recorded evaluation configuration: {"Segmentation Level":"middle (2)"}. This record does not assert independent reproduction.

### PointNet

Shape mIoU: Avg: 50.1. Result ID: 698ca09c0b614d12cff20f9c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634f1e209131ce9fe019.md>)

Recorded evaluation configuration: {"Segmentation Level":"middle (2)"}. This record does not assert independent reproduction.

### PointCNN

Shape mIoU: Avg: 49.6. Result ID: 698ca09c0b614d12cff20fa6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

Recorded evaluation configuration: {"Segmentation Level":"fine-grained (3)"}. This record does not assert independent reproduction.

### PointNet

Shape mIoU: Avg: 48.2. Result ID: 698ca09c0b614d12cff20f9d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634f1e209131ce9fe019.md>)

Recorded evaluation configuration: {"Segmentation Level":"fine-grained (3)"}. This record does not assert independent reproduction.

### SpiderCNN

Shape mIoU: Avg: 48. Result ID: 698ca09c0b614d12cff20fa3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa65301e209131cea08bc3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64d91e209131cea06a9f.md>)

Recorded evaluation configuration: {"Segmentation Level":"fine-grained (3)"}. This record does not assert independent reproduction.

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