Submitted by ZEYU HU.

Submission data

Full nameVoxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation
DescriptionA novel 3D deep architecture operating on the voxel and mesh representations leveraging both the euclidean and geodesic information.
Publication titleVMNet: Voxel-Mesh Network for Geodesic-Aware 3D Semantic Segmentation
Publication authorsZeyu HU, Xuyang Bai, Jiaxiang Shang, Runze Zhang, Jiayu Dong, Xin Wang, Guangyuan Sun, Hongbo Fu, Chiew-Lan Tai
Publication venueICCV 2021 (Oral)
Publication URLhttps://arxiv.org/abs/2107.13824
Input Data TypesUses Color        Uses 3D
Programming language(s)Python with Pytorch
HardwareTesla V100
Websitehttps://github.com/hzykent/VMNet
Source code or download URLhttps://github.com/hzykent/VMNet
Submission creation date18 Jan, 2021
Last edited30 Jul, 2021

3D semantic label results

Infoavg ioubathtubbedbookshelfcabinetchaircountercurtaindeskdoorfloorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwallwindow
permissive0.7460.8700.8380.8580.7290.8500.5010.8740.5870.6580.9560.5640.2990.7650.9000.7160.8120.6310.9390.8580.709