Submitted by Xiaoyang Wu.

Submission data

Full namePoint Transformer V2
Publication titlePoint Transformer V2: Grouped Vector Attention and Partition-based Pooling
Publication authorsXiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao
Publication venueNeurIPS 2022
Publication URLhttps://arxiv.org/abs/2210.05666
Input Data TypesUses Color        Uses 3D
Programming language(s)Python
HardwareGeForce 3090
Websitehttps://github.com/Pointcept/PointTransformerV2
Source code or download URLhttps://github.com/Pointcept/PointTransformerV2
Submission creation date1 Aug, 2022
Last edited16 Jan, 2024

3D semantic label results

Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
0.7520.7420.8090.8720.7580.8600.5520.8910.6100.6870.9600.5590.3040.7660.9260.7670.7970.6440.9420.8760.722