# 3D Semantic Segmentation on S3DIS (all six areas) Source: [3D Semantic Segmentation on S3DIS (all six areas)]() Collection index: [llms.txt]() Data updated: 2026-02-27T06:37:59.254Z Evaluation of 3D Semantic Segmentation performance across all six areas of the Stanford Indoor Dataset (S3DIS). 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 Semantic Segmentation | | Dataset (including any stated split) | S3DIS 1 (all six areas) | | Metric | Overall Accuracy (OA) | | 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 - [Overall Accuracy (OA)](): Higher is better; unit: PERCENT. - [Mean IoU (mIoU)](): Higher is better; unit: PERCENT. - [IoU (Ceiling)](): Higher is better; unit: PERCENT. - [IoU (Floor)](): Higher is better; unit: PERCENT. - [IoU (Wall)](): Higher is better; unit: PERCENT. - [IoU (Beam)](): Higher is better; unit: PERCENT. - [IoU (Column)](): Higher is better; unit: PERCENT. - [IoU (Window)](): Higher is better; unit: PERCENT. - [IoU (Door)](): Higher is better; unit: PERCENT. - [IoU (Chair)](): Higher is better; unit: PERCENT. - [IoU (Table)](): Higher is better; unit: PERCENT. - [IoU (Bookcase)](): Higher is better; unit: PERCENT. - [IoU (Sofa)](): Higher is better; unit: PERCENT. - [IoU (Board)](): Higher is better; unit: PERCENT. - [IoU (Clutter)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Overall Accuracy (OA) | Result date | Report paper | | --- | --- | --- | --- | | MuGNet | 88.5 | Not recorded | 68fa692e1e209131cea209c2 | | RandLA-Net | 88 | Not recorded | 68fa692e1e209131cea209c2 | | SPG | 85.5 | Not recorded | 68fa692e1e209131cea209c2 | | DGCNN | 84.1 | Not recorded | 68fa692e1e209131cea209c2 | | Engelmann et al. | 81.1 | Not recorded | 68fa692e1e209131cea209c2 | | PointNet | 78.5 | Not recorded | 68fa692e1e209131cea209c2 | ### MuGNet Overall Accuracy (OA): 88.5. Result ID: 69a13bad35dfcabb00e5a8c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RandLA-Net Overall Accuracy (OA): 88. Result ID: 69a13bad35dfcabb00e5a8bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SPG Overall Accuracy (OA): 85.5. Result ID: 69a13bad35dfcabb00e5a8be. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DGCNN Overall Accuracy (OA): 84.1. Result ID: 69a13bad35dfcabb00e5a8bd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Engelmann et al. Overall Accuracy (OA): 81.1. Result ID: 69a13bad35dfcabb00e5a8bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### PointNet Overall Accuracy (OA): 78.5. Result ID: 69a13bad35dfcabb00e5a8bb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.