Submitted by Xiaoyang Wu.

Submission data

Full namePoint Prompt Training
Publication titleTowards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training
Publication authorsXiaoyang Wu, Zhuotao Tian, Xin Wen, Bohao Peng, Xihui Liu, Kaicheng Yu, Hengshuang Zhao
Publication venueCVPR 2024
Publication URLhttps://arxiv.org/abs/2308.09718
Input Data TypesUses Color        Uses 3D
Programming language(s)Python, C++ with CUDA
HardwareNVIDIA A100
Websitehttps://github.com/Pointcept/Pointcept
Source code or download URLhttps://github.com/Pointcept/Pointcept
Submission creation date28 Feb, 2023
Last edited2 Mar, 2024

3D semantic label results

Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
0.7660.9320.7940.8290.7510.8540.5400.9030.6300.6720.9630.5650.3570.7880.9000.7370.8020.6850.9500.8870.780