Submitted anonymously.

### Submission data

 Full name SoftGroup++: Scalable 3D Instance Segmentation with Octree Pyramid Grouping Description SoftGroup++ is built upon SoftGroup, which differs in three important aspects: (1) performs octree $k$-NN instead of vanilla $k$-NN to reduce time complexity from $\mathcal{O}(n^2)$ to $\mathcal{O}(n \log n)$, (2) performs pyramid scaling that adaptively downsample backbone outputs to reduce search space for $k$-NN and grouping modules, and (3) performs late devoxelization that delay the conversion from voxel to points towards the end of the model such that intermediate components runs at low computational cost. Extensive experiments on various indoor and outdoor datasets demonstrate the efficacy of the proposed SoftGroup++. Notably, SoftGroup++ processes large scenes of millions points by a single forward without dividing the input into multiple parts, thus providing training-inference consistency. Especially, SoftGroup++ achieves 2.4 points AP$_{50}$ improvement while nearly $6\times$ faster than existing fastest method on S3DIS dataset. Input Data Types Uses Color,Uses Geometry        Uses 3D Programming language(s) Python, Cuda Hardware RTX 8000 Submission creation date 27 Apr, 2022 Last edited 19 May, 2022

### 3D semantic instance results

Infoavg ap 50%bathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
0.7691.0000.8030.9370.6840.8650.2130.8700.6640.5710.7580.7020.8071.0000.6530.9020.7921.0000.626